Upload folder using huggingface_hub
Browse files- checkpoint-31350/config.json +25 -0
- checkpoint-31350/generation_config.json +5 -0
- checkpoint-31350/model.safetensors +3 -0
- checkpoint-31350/optimizer.pt +3 -0
- checkpoint-31350/rng_state.pth +3 -0
- checkpoint-31350/scheduler.pt +3 -0
- checkpoint-31350/special_tokens_map.json +37 -0
- checkpoint-31350/tokenizer.json +0 -0
- checkpoint-31350/tokenizer_config.json +56 -0
- checkpoint-31350/trainer_state.json +2264 -0
- checkpoint-31350/training_args.bin +3 -0
- checkpoint-31350/vocab.txt +0 -0
- checkpoint-37620/config.json +25 -0
- checkpoint-37620/generation_config.json +5 -0
- checkpoint-37620/model.safetensors +3 -0
- checkpoint-37620/optimizer.pt +3 -0
- checkpoint-37620/rng_state.pth +3 -0
- checkpoint-37620/scheduler.pt +3 -0
- checkpoint-37620/special_tokens_map.json +37 -0
- checkpoint-37620/tokenizer.json +0 -0
- checkpoint-37620/tokenizer_config.json +56 -0
- checkpoint-37620/trainer_state.json +2713 -0
- checkpoint-37620/training_args.bin +3 -0
- checkpoint-37620/vocab.txt +0 -0
- config.json +25 -0
- generation_config.json +5 -0
- model.safetensors +3 -0
- special_tokens_map.json +37 -0
- tokenizer.json +0 -0
- tokenizer_config.json +56 -0
- vocab.txt +0 -0
checkpoint-31350/config.json
ADDED
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "Ransaka/sinhala-bert-medium-v2",
|
3 |
+
"architectures": [
|
4 |
+
"BertForMaskedLM"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_dropout": null,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_dropout_prob": 0.1,
|
10 |
+
"hidden_size": 786,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 1024,
|
13 |
+
"layer_norm_eps": 1e-12,
|
14 |
+
"max_position_embeddings": 256,
|
15 |
+
"model_type": "bert",
|
16 |
+
"num_attention_heads": 6,
|
17 |
+
"num_hidden_layers": 6,
|
18 |
+
"pad_token_id": 0,
|
19 |
+
"position_embedding_type": "absolute",
|
20 |
+
"torch_dtype": "float32",
|
21 |
+
"transformers_version": "4.47.0",
|
22 |
+
"type_vocab_size": 2,
|
23 |
+
"use_cache": true,
|
24 |
+
"vocab_size": 32000
|
25 |
+
}
|
checkpoint-31350/generation_config.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"pad_token_id": 0,
|
4 |
+
"transformers_version": "4.47.0"
|
5 |
+
}
|
checkpoint-31350/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1ad0817b38f7a067e234a9c4ac9a1483e59d73100b5cd4684e280981127e5065
|
3 |
+
size 202182680
|
checkpoint-31350/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:58f50107cab8c8fe502d465b6609db1651d64a1d8ebcf060f5969ca573f8a2c8
|
3 |
+
size 404427706
|
checkpoint-31350/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f33e34f57c3dfd177ffc0b61ce010f0da3fa1a2f4a131f755d497f5b5415949d
|
3 |
+
size 14244
|
checkpoint-31350/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:29fdc16fbaa13005c89edbafc02c1ba8a1a12e61e60e8bea63235eca8e56a2ab
|
3 |
+
size 1064
|
checkpoint-31350/special_tokens_map.json
ADDED
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": {
|
3 |
+
"content": "[CLS]",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"mask_token": {
|
10 |
+
"content": "[MASK]",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"pad_token": {
|
17 |
+
"content": "[PAD]",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"sep_token": {
|
24 |
+
"content": "[SEP]",
|
25 |
+
"lstrip": false,
|
26 |
+
"normalized": false,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
},
|
30 |
+
"unk_token": {
|
31 |
+
"content": "[UNK]",
|
32 |
+
"lstrip": false,
|
33 |
+
"normalized": false,
|
34 |
+
"rstrip": false,
|
35 |
+
"single_word": false
|
36 |
+
}
|
37 |
+
}
|
checkpoint-31350/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-31350/tokenizer_config.json
ADDED
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "[PAD]",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "[UNK]",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "[CLS]",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "[SEP]",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"4": {
|
36 |
+
"content": "[MASK]",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"clean_up_tokenization_spaces": true,
|
45 |
+
"cls_token": "[CLS]",
|
46 |
+
"do_lower_case": false,
|
47 |
+
"extra_special_tokens": {},
|
48 |
+
"mask_token": "[MASK]",
|
49 |
+
"model_max_length": 512,
|
50 |
+
"pad_token": "[PAD]",
|
51 |
+
"sep_token": "[SEP]",
|
52 |
+
"strip_accents": null,
|
53 |
+
"tokenize_chinese_chars": true,
|
54 |
+
"tokenizer_class": "BertTokenizer",
|
55 |
+
"unk_token": "[UNK]"
|
56 |
+
}
|
checkpoint-31350/trainer_state.json
ADDED
@@ -0,0 +1,2264 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 5.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 31350,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.01594896331738437,
|
13 |
+
"grad_norm": 793786.1875,
|
14 |
+
"learning_rate": 4.9867091972355135e-05,
|
15 |
+
"loss": 4.3231,
|
16 |
+
"step": 100
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.03189792663476874,
|
20 |
+
"grad_norm": 705751.0625,
|
21 |
+
"learning_rate": 4.973418394471026e-05,
|
22 |
+
"loss": 4.1867,
|
23 |
+
"step": 200
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.04784688995215311,
|
27 |
+
"grad_norm": 761077.1875,
|
28 |
+
"learning_rate": 4.9601275917065395e-05,
|
29 |
+
"loss": 4.225,
|
30 |
+
"step": 300
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.06379585326953748,
|
34 |
+
"grad_norm": 660073.9375,
|
35 |
+
"learning_rate": 4.946836788942053e-05,
|
36 |
+
"loss": 4.2529,
|
37 |
+
"step": 400
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.07974481658692185,
|
41 |
+
"grad_norm": 701404.4375,
|
42 |
+
"learning_rate": 4.9335459861775654e-05,
|
43 |
+
"loss": 4.2985,
|
44 |
+
"step": 500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.09569377990430622,
|
48 |
+
"grad_norm": 776949.5625,
|
49 |
+
"learning_rate": 4.920255183413078e-05,
|
50 |
+
"loss": 4.2124,
|
51 |
+
"step": 600
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.11164274322169059,
|
55 |
+
"grad_norm": 703487.375,
|
56 |
+
"learning_rate": 4.906964380648591e-05,
|
57 |
+
"loss": 4.224,
|
58 |
+
"step": 700
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.12759170653907495,
|
62 |
+
"grad_norm": 727441.625,
|
63 |
+
"learning_rate": 4.893673577884104e-05,
|
64 |
+
"loss": 4.3259,
|
65 |
+
"step": 800
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.14354066985645933,
|
69 |
+
"grad_norm": 772143.4375,
|
70 |
+
"learning_rate": 4.880382775119617e-05,
|
71 |
+
"loss": 4.1641,
|
72 |
+
"step": 900
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.1594896331738437,
|
76 |
+
"grad_norm": 834641.6875,
|
77 |
+
"learning_rate": 4.8670919723551306e-05,
|
78 |
+
"loss": 4.2651,
|
79 |
+
"step": 1000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.17543859649122806,
|
83 |
+
"grad_norm": 743223.0,
|
84 |
+
"learning_rate": 4.853801169590643e-05,
|
85 |
+
"loss": 4.1867,
|
86 |
+
"step": 1100
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.19138755980861244,
|
90 |
+
"grad_norm": 643478.125,
|
91 |
+
"learning_rate": 4.8405103668261565e-05,
|
92 |
+
"loss": 4.1816,
|
93 |
+
"step": 1200
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.20733652312599682,
|
97 |
+
"grad_norm": 656405.6875,
|
98 |
+
"learning_rate": 4.82721956406167e-05,
|
99 |
+
"loss": 4.1677,
|
100 |
+
"step": 1300
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.22328548644338117,
|
104 |
+
"grad_norm": 714701.3125,
|
105 |
+
"learning_rate": 4.813928761297183e-05,
|
106 |
+
"loss": 4.1595,
|
107 |
+
"step": 1400
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.23923444976076555,
|
111 |
+
"grad_norm": 739746.875,
|
112 |
+
"learning_rate": 4.800637958532696e-05,
|
113 |
+
"loss": 4.1319,
|
114 |
+
"step": 1500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.2551834130781499,
|
118 |
+
"grad_norm": 736049.3125,
|
119 |
+
"learning_rate": 4.787347155768209e-05,
|
120 |
+
"loss": 4.1543,
|
121 |
+
"step": 1600
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.2711323763955343,
|
125 |
+
"grad_norm": 668736.5,
|
126 |
+
"learning_rate": 4.7740563530037217e-05,
|
127 |
+
"loss": 4.2052,
|
128 |
+
"step": 1700
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.28708133971291866,
|
132 |
+
"grad_norm": 655993.0625,
|
133 |
+
"learning_rate": 4.760765550239234e-05,
|
134 |
+
"loss": 4.1642,
|
135 |
+
"step": 1800
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.30303030303030304,
|
139 |
+
"grad_norm": 732162.3125,
|
140 |
+
"learning_rate": 4.7474747474747476e-05,
|
141 |
+
"loss": 4.1464,
|
142 |
+
"step": 1900
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.3189792663476874,
|
146 |
+
"grad_norm": 687130.875,
|
147 |
+
"learning_rate": 4.734183944710261e-05,
|
148 |
+
"loss": 4.181,
|
149 |
+
"step": 2000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.3349282296650718,
|
153 |
+
"grad_norm": 823373.1875,
|
154 |
+
"learning_rate": 4.7208931419457735e-05,
|
155 |
+
"loss": 4.1591,
|
156 |
+
"step": 2100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.3508771929824561,
|
160 |
+
"grad_norm": 716504.625,
|
161 |
+
"learning_rate": 4.707602339181287e-05,
|
162 |
+
"loss": 4.1391,
|
163 |
+
"step": 2200
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.3668261562998405,
|
167 |
+
"grad_norm": 714200.6875,
|
168 |
+
"learning_rate": 4.6943115364168e-05,
|
169 |
+
"loss": 4.0962,
|
170 |
+
"step": 2300
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.3827751196172249,
|
174 |
+
"grad_norm": 662553.0,
|
175 |
+
"learning_rate": 4.681020733652313e-05,
|
176 |
+
"loss": 4.0964,
|
177 |
+
"step": 2400
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.39872408293460926,
|
181 |
+
"grad_norm": 767714.875,
|
182 |
+
"learning_rate": 4.667729930887826e-05,
|
183 |
+
"loss": 4.1174,
|
184 |
+
"step": 2500
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.41467304625199364,
|
188 |
+
"grad_norm": 763852.75,
|
189 |
+
"learning_rate": 4.6544391281233393e-05,
|
190 |
+
"loss": 4.0442,
|
191 |
+
"step": 2600
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.430622009569378,
|
195 |
+
"grad_norm": 733045.8125,
|
196 |
+
"learning_rate": 4.641148325358852e-05,
|
197 |
+
"loss": 4.112,
|
198 |
+
"step": 2700
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.44657097288676234,
|
202 |
+
"grad_norm": 752065.875,
|
203 |
+
"learning_rate": 4.6278575225943646e-05,
|
204 |
+
"loss": 4.1223,
|
205 |
+
"step": 2800
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.4625199362041467,
|
209 |
+
"grad_norm": 675739.8125,
|
210 |
+
"learning_rate": 4.614566719829878e-05,
|
211 |
+
"loss": 4.1608,
|
212 |
+
"step": 2900
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.4784688995215311,
|
216 |
+
"grad_norm": 666147.375,
|
217 |
+
"learning_rate": 4.6012759170653905e-05,
|
218 |
+
"loss": 4.0803,
|
219 |
+
"step": 3000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.4944178628389155,
|
223 |
+
"grad_norm": 594287.9375,
|
224 |
+
"learning_rate": 4.587985114300904e-05,
|
225 |
+
"loss": 4.045,
|
226 |
+
"step": 3100
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.5103668261562998,
|
230 |
+
"grad_norm": 778230.25,
|
231 |
+
"learning_rate": 4.574694311536417e-05,
|
232 |
+
"loss": 4.1268,
|
233 |
+
"step": 3200
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.5263157894736842,
|
237 |
+
"grad_norm": 683312.0625,
|
238 |
+
"learning_rate": 4.56140350877193e-05,
|
239 |
+
"loss": 4.0424,
|
240 |
+
"step": 3300
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.5422647527910686,
|
244 |
+
"grad_norm": 640858.25,
|
245 |
+
"learning_rate": 4.548112706007443e-05,
|
246 |
+
"loss": 4.0235,
|
247 |
+
"step": 3400
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.5582137161084529,
|
251 |
+
"grad_norm": 680217.875,
|
252 |
+
"learning_rate": 4.5348219032429564e-05,
|
253 |
+
"loss": 4.1423,
|
254 |
+
"step": 3500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.5741626794258373,
|
258 |
+
"grad_norm": 708408.4375,
|
259 |
+
"learning_rate": 4.521531100478469e-05,
|
260 |
+
"loss": 4.0291,
|
261 |
+
"step": 3600
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.5901116427432217,
|
265 |
+
"grad_norm": 663333.0,
|
266 |
+
"learning_rate": 4.508240297713982e-05,
|
267 |
+
"loss": 3.9998,
|
268 |
+
"step": 3700
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.6060606060606061,
|
272 |
+
"grad_norm": 652592.0,
|
273 |
+
"learning_rate": 4.494949494949495e-05,
|
274 |
+
"loss": 4.0424,
|
275 |
+
"step": 3800
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.6220095693779905,
|
279 |
+
"grad_norm": 667416.9375,
|
280 |
+
"learning_rate": 4.481658692185008e-05,
|
281 |
+
"loss": 4.053,
|
282 |
+
"step": 3900
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.6379585326953748,
|
286 |
+
"grad_norm": 674108.4375,
|
287 |
+
"learning_rate": 4.468367889420521e-05,
|
288 |
+
"loss": 4.0748,
|
289 |
+
"step": 4000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.6539074960127592,
|
293 |
+
"grad_norm": 714003.625,
|
294 |
+
"learning_rate": 4.455077086656034e-05,
|
295 |
+
"loss": 3.9807,
|
296 |
+
"step": 4100
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.6698564593301436,
|
300 |
+
"grad_norm": 693166.4375,
|
301 |
+
"learning_rate": 4.4417862838915475e-05,
|
302 |
+
"loss": 3.993,
|
303 |
+
"step": 4200
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.6858054226475279,
|
307 |
+
"grad_norm": 689997.125,
|
308 |
+
"learning_rate": 4.42849548112706e-05,
|
309 |
+
"loss": 4.0593,
|
310 |
+
"step": 4300
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.7017543859649122,
|
314 |
+
"grad_norm": 741118.5625,
|
315 |
+
"learning_rate": 4.4152046783625734e-05,
|
316 |
+
"loss": 4.0409,
|
317 |
+
"step": 4400
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.7177033492822966,
|
321 |
+
"grad_norm": 668626.9375,
|
322 |
+
"learning_rate": 4.401913875598087e-05,
|
323 |
+
"loss": 4.06,
|
324 |
+
"step": 4500
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.733652312599681,
|
328 |
+
"grad_norm": 771064.125,
|
329 |
+
"learning_rate": 4.388623072833599e-05,
|
330 |
+
"loss": 4.0122,
|
331 |
+
"step": 4600
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.7496012759170654,
|
335 |
+
"grad_norm": 756131.375,
|
336 |
+
"learning_rate": 4.3753322700691126e-05,
|
337 |
+
"loss": 4.043,
|
338 |
+
"step": 4700
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.7655502392344498,
|
342 |
+
"grad_norm": 737817.5625,
|
343 |
+
"learning_rate": 4.362041467304626e-05,
|
344 |
+
"loss": 4.0213,
|
345 |
+
"step": 4800
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.7814992025518341,
|
349 |
+
"grad_norm": 758023.5625,
|
350 |
+
"learning_rate": 4.3487506645401385e-05,
|
351 |
+
"loss": 3.9778,
|
352 |
+
"step": 4900
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.7974481658692185,
|
356 |
+
"grad_norm": 672895.1875,
|
357 |
+
"learning_rate": 4.335459861775651e-05,
|
358 |
+
"loss": 4.0434,
|
359 |
+
"step": 5000
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.8133971291866029,
|
363 |
+
"grad_norm": 710937.375,
|
364 |
+
"learning_rate": 4.3221690590111645e-05,
|
365 |
+
"loss": 4.0117,
|
366 |
+
"step": 5100
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.8293460925039873,
|
370 |
+
"grad_norm": 773563.875,
|
371 |
+
"learning_rate": 4.308878256246677e-05,
|
372 |
+
"loss": 4.0516,
|
373 |
+
"step": 5200
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.8452950558213717,
|
377 |
+
"grad_norm": 750733.4375,
|
378 |
+
"learning_rate": 4.2955874534821904e-05,
|
379 |
+
"loss": 3.9798,
|
380 |
+
"step": 5300
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.861244019138756,
|
384 |
+
"grad_norm": 706317.3125,
|
385 |
+
"learning_rate": 4.282296650717704e-05,
|
386 |
+
"loss": 3.9845,
|
387 |
+
"step": 5400
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.8771929824561403,
|
391 |
+
"grad_norm": 710855.75,
|
392 |
+
"learning_rate": 4.269005847953216e-05,
|
393 |
+
"loss": 4.0295,
|
394 |
+
"step": 5500
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.8931419457735247,
|
398 |
+
"grad_norm": 712988.9375,
|
399 |
+
"learning_rate": 4.2557150451887296e-05,
|
400 |
+
"loss": 3.9732,
|
401 |
+
"step": 5600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.9090909090909091,
|
405 |
+
"grad_norm": 628492.75,
|
406 |
+
"learning_rate": 4.242424242424243e-05,
|
407 |
+
"loss": 4.0012,
|
408 |
+
"step": 5700
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.9250398724082934,
|
412 |
+
"grad_norm": 821738.375,
|
413 |
+
"learning_rate": 4.2291334396597556e-05,
|
414 |
+
"loss": 3.9867,
|
415 |
+
"step": 5800
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.9409888357256778,
|
419 |
+
"grad_norm": 720818.125,
|
420 |
+
"learning_rate": 4.215842636895269e-05,
|
421 |
+
"loss": 3.962,
|
422 |
+
"step": 5900
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.9569377990430622,
|
426 |
+
"grad_norm": 698428.0625,
|
427 |
+
"learning_rate": 4.2025518341307815e-05,
|
428 |
+
"loss": 4.0095,
|
429 |
+
"step": 6000
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.9728867623604466,
|
433 |
+
"grad_norm": 769185.75,
|
434 |
+
"learning_rate": 4.189261031366295e-05,
|
435 |
+
"loss": 3.9469,
|
436 |
+
"step": 6100
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.988835725677831,
|
440 |
+
"grad_norm": 706422.25,
|
441 |
+
"learning_rate": 4.1759702286018074e-05,
|
442 |
+
"loss": 3.9466,
|
443 |
+
"step": 6200
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 1.0,
|
447 |
+
"eval_loss": 3.8981289863586426,
|
448 |
+
"eval_runtime": 213.136,
|
449 |
+
"eval_samples_per_second": 134.295,
|
450 |
+
"eval_steps_per_second": 4.199,
|
451 |
+
"step": 6270
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 1.0047846889952152,
|
455 |
+
"grad_norm": 664245.5,
|
456 |
+
"learning_rate": 4.162679425837321e-05,
|
457 |
+
"loss": 3.9397,
|
458 |
+
"step": 6300
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 1.0207336523125996,
|
462 |
+
"grad_norm": 790251.5625,
|
463 |
+
"learning_rate": 4.149388623072834e-05,
|
464 |
+
"loss": 3.7918,
|
465 |
+
"step": 6400
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 1.036682615629984,
|
469 |
+
"grad_norm": 813129.1875,
|
470 |
+
"learning_rate": 4.1360978203083467e-05,
|
471 |
+
"loss": 3.8563,
|
472 |
+
"step": 6500
|
473 |
+
},
|
474 |
+
{
|
475 |
+
"epoch": 1.0526315789473684,
|
476 |
+
"grad_norm": 695871.5,
|
477 |
+
"learning_rate": 4.12280701754386e-05,
|
478 |
+
"loss": 3.959,
|
479 |
+
"step": 6600
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 1.0685805422647527,
|
483 |
+
"grad_norm": 640218.3125,
|
484 |
+
"learning_rate": 4.109516214779373e-05,
|
485 |
+
"loss": 3.9882,
|
486 |
+
"step": 6700
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 1.0845295055821371,
|
490 |
+
"grad_norm": 709886.125,
|
491 |
+
"learning_rate": 4.096225412014886e-05,
|
492 |
+
"loss": 3.9328,
|
493 |
+
"step": 6800
|
494 |
+
},
|
495 |
+
{
|
496 |
+
"epoch": 1.1004784688995215,
|
497 |
+
"grad_norm": 663732.1875,
|
498 |
+
"learning_rate": 4.082934609250399e-05,
|
499 |
+
"loss": 3.947,
|
500 |
+
"step": 6900
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 1.1164274322169059,
|
504 |
+
"grad_norm": 649781.9375,
|
505 |
+
"learning_rate": 4.069643806485912e-05,
|
506 |
+
"loss": 3.8735,
|
507 |
+
"step": 7000
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 1.1323763955342903,
|
511 |
+
"grad_norm": 788934.4375,
|
512 |
+
"learning_rate": 4.056353003721425e-05,
|
513 |
+
"loss": 3.8548,
|
514 |
+
"step": 7100
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 1.1483253588516746,
|
518 |
+
"grad_norm": 688279.3125,
|
519 |
+
"learning_rate": 4.043062200956938e-05,
|
520 |
+
"loss": 3.8844,
|
521 |
+
"step": 7200
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 1.164274322169059,
|
525 |
+
"grad_norm": 698958.0,
|
526 |
+
"learning_rate": 4.029771398192451e-05,
|
527 |
+
"loss": 3.878,
|
528 |
+
"step": 7300
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 1.1802232854864434,
|
532 |
+
"grad_norm": 777544.8125,
|
533 |
+
"learning_rate": 4.016480595427964e-05,
|
534 |
+
"loss": 3.8299,
|
535 |
+
"step": 7400
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 1.1961722488038278,
|
539 |
+
"grad_norm": 768888.3125,
|
540 |
+
"learning_rate": 4.003189792663477e-05,
|
541 |
+
"loss": 3.859,
|
542 |
+
"step": 7500
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 1.2121212121212122,
|
546 |
+
"grad_norm": 703599.8125,
|
547 |
+
"learning_rate": 3.98989898989899e-05,
|
548 |
+
"loss": 3.9008,
|
549 |
+
"step": 7600
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 1.2280701754385965,
|
553 |
+
"grad_norm": 684616.75,
|
554 |
+
"learning_rate": 3.976608187134503e-05,
|
555 |
+
"loss": 3.8917,
|
556 |
+
"step": 7700
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 1.244019138755981,
|
560 |
+
"grad_norm": 754650.9375,
|
561 |
+
"learning_rate": 3.963317384370016e-05,
|
562 |
+
"loss": 3.7987,
|
563 |
+
"step": 7800
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 1.2599681020733653,
|
567 |
+
"grad_norm": 833289.8125,
|
568 |
+
"learning_rate": 3.9500265816055295e-05,
|
569 |
+
"loss": 3.8097,
|
570 |
+
"step": 7900
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 1.2759170653907497,
|
574 |
+
"grad_norm": 712939.125,
|
575 |
+
"learning_rate": 3.936735778841042e-05,
|
576 |
+
"loss": 3.91,
|
577 |
+
"step": 8000
|
578 |
+
},
|
579 |
+
{
|
580 |
+
"epoch": 1.291866028708134,
|
581 |
+
"grad_norm": 691034.4375,
|
582 |
+
"learning_rate": 3.9234449760765554e-05,
|
583 |
+
"loss": 3.8607,
|
584 |
+
"step": 8100
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 1.3078149920255182,
|
588 |
+
"grad_norm": 777656.1875,
|
589 |
+
"learning_rate": 3.910154173312068e-05,
|
590 |
+
"loss": 3.8294,
|
591 |
+
"step": 8200
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 1.3237639553429026,
|
595 |
+
"grad_norm": 779268.6875,
|
596 |
+
"learning_rate": 3.896863370547581e-05,
|
597 |
+
"loss": 3.839,
|
598 |
+
"step": 8300
|
599 |
+
},
|
600 |
+
{
|
601 |
+
"epoch": 1.339712918660287,
|
602 |
+
"grad_norm": 737519.0,
|
603 |
+
"learning_rate": 3.883572567783094e-05,
|
604 |
+
"loss": 3.8773,
|
605 |
+
"step": 8400
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 1.3556618819776713,
|
609 |
+
"grad_norm": 722279.3125,
|
610 |
+
"learning_rate": 3.870281765018607e-05,
|
611 |
+
"loss": 3.8351,
|
612 |
+
"step": 8500
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 1.3716108452950557,
|
616 |
+
"grad_norm": 763525.9375,
|
617 |
+
"learning_rate": 3.8569909622541206e-05,
|
618 |
+
"loss": 3.8517,
|
619 |
+
"step": 8600
|
620 |
+
},
|
621 |
+
{
|
622 |
+
"epoch": 1.38755980861244,
|
623 |
+
"grad_norm": 715289.3125,
|
624 |
+
"learning_rate": 3.843700159489633e-05,
|
625 |
+
"loss": 3.8272,
|
626 |
+
"step": 8700
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 1.4035087719298245,
|
630 |
+
"grad_norm": 656186.625,
|
631 |
+
"learning_rate": 3.8304093567251465e-05,
|
632 |
+
"loss": 3.8894,
|
633 |
+
"step": 8800
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 1.4194577352472089,
|
637 |
+
"grad_norm": 756528.875,
|
638 |
+
"learning_rate": 3.81711855396066e-05,
|
639 |
+
"loss": 3.8387,
|
640 |
+
"step": 8900
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 1.4354066985645932,
|
644 |
+
"grad_norm": 716015.0,
|
645 |
+
"learning_rate": 3.8038277511961725e-05,
|
646 |
+
"loss": 3.7842,
|
647 |
+
"step": 9000
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 1.4513556618819776,
|
651 |
+
"grad_norm": 721006.3125,
|
652 |
+
"learning_rate": 3.790536948431686e-05,
|
653 |
+
"loss": 3.7769,
|
654 |
+
"step": 9100
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 1.467304625199362,
|
658 |
+
"grad_norm": 800538.1875,
|
659 |
+
"learning_rate": 3.7772461456671984e-05,
|
660 |
+
"loss": 3.8393,
|
661 |
+
"step": 9200
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 1.4832535885167464,
|
665 |
+
"grad_norm": 699156.0,
|
666 |
+
"learning_rate": 3.763955342902711e-05,
|
667 |
+
"loss": 3.7527,
|
668 |
+
"step": 9300
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 1.4992025518341308,
|
672 |
+
"grad_norm": 699306.375,
|
673 |
+
"learning_rate": 3.750664540138224e-05,
|
674 |
+
"loss": 3.8306,
|
675 |
+
"step": 9400
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 1.5151515151515151,
|
679 |
+
"grad_norm": 689892.5,
|
680 |
+
"learning_rate": 3.7373737373737376e-05,
|
681 |
+
"loss": 3.8582,
|
682 |
+
"step": 9500
|
683 |
+
},
|
684 |
+
{
|
685 |
+
"epoch": 1.5311004784688995,
|
686 |
+
"grad_norm": 712134.3125,
|
687 |
+
"learning_rate": 3.72408293460925e-05,
|
688 |
+
"loss": 3.9048,
|
689 |
+
"step": 9600
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 1.547049441786284,
|
693 |
+
"grad_norm": 611158.625,
|
694 |
+
"learning_rate": 3.7107921318447635e-05,
|
695 |
+
"loss": 3.8315,
|
696 |
+
"step": 9700
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 1.5629984051036683,
|
700 |
+
"grad_norm": 814951.875,
|
701 |
+
"learning_rate": 3.697501329080277e-05,
|
702 |
+
"loss": 3.8433,
|
703 |
+
"step": 9800
|
704 |
+
},
|
705 |
+
{
|
706 |
+
"epoch": 1.5789473684210527,
|
707 |
+
"grad_norm": 638232.25,
|
708 |
+
"learning_rate": 3.6842105263157895e-05,
|
709 |
+
"loss": 3.8697,
|
710 |
+
"step": 9900
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 1.594896331738437,
|
714 |
+
"grad_norm": 717277.8125,
|
715 |
+
"learning_rate": 3.670919723551303e-05,
|
716 |
+
"loss": 3.8447,
|
717 |
+
"step": 10000
|
718 |
+
},
|
719 |
+
{
|
720 |
+
"epoch": 1.6108452950558214,
|
721 |
+
"grad_norm": 623000.1875,
|
722 |
+
"learning_rate": 3.657628920786816e-05,
|
723 |
+
"loss": 3.9104,
|
724 |
+
"step": 10100
|
725 |
+
},
|
726 |
+
{
|
727 |
+
"epoch": 1.6267942583732058,
|
728 |
+
"grad_norm": 737050.875,
|
729 |
+
"learning_rate": 3.644338118022329e-05,
|
730 |
+
"loss": 3.8208,
|
731 |
+
"step": 10200
|
732 |
+
},
|
733 |
+
{
|
734 |
+
"epoch": 1.6427432216905902,
|
735 |
+
"grad_norm": 766050.0,
|
736 |
+
"learning_rate": 3.631047315257842e-05,
|
737 |
+
"loss": 3.8454,
|
738 |
+
"step": 10300
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 1.6586921850079746,
|
742 |
+
"grad_norm": 681873.25,
|
743 |
+
"learning_rate": 3.6177565124933546e-05,
|
744 |
+
"loss": 3.8284,
|
745 |
+
"step": 10400
|
746 |
+
},
|
747 |
+
{
|
748 |
+
"epoch": 1.674641148325359,
|
749 |
+
"grad_norm": 636321.0625,
|
750 |
+
"learning_rate": 3.604465709728867e-05,
|
751 |
+
"loss": 3.8191,
|
752 |
+
"step": 10500
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 1.6905901116427433,
|
756 |
+
"grad_norm": 751694.875,
|
757 |
+
"learning_rate": 3.5911749069643806e-05,
|
758 |
+
"loss": 3.8152,
|
759 |
+
"step": 10600
|
760 |
+
},
|
761 |
+
{
|
762 |
+
"epoch": 1.7065390749601277,
|
763 |
+
"grad_norm": 779390.25,
|
764 |
+
"learning_rate": 3.577884104199894e-05,
|
765 |
+
"loss": 3.8235,
|
766 |
+
"step": 10700
|
767 |
+
},
|
768 |
+
{
|
769 |
+
"epoch": 1.722488038277512,
|
770 |
+
"grad_norm": 780193.4375,
|
771 |
+
"learning_rate": 3.5645933014354065e-05,
|
772 |
+
"loss": 3.8109,
|
773 |
+
"step": 10800
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 1.7384370015948964,
|
777 |
+
"grad_norm": 678532.75,
|
778 |
+
"learning_rate": 3.55130249867092e-05,
|
779 |
+
"loss": 3.8017,
|
780 |
+
"step": 10900
|
781 |
+
},
|
782 |
+
{
|
783 |
+
"epoch": 1.7543859649122808,
|
784 |
+
"grad_norm": 733045.375,
|
785 |
+
"learning_rate": 3.538011695906433e-05,
|
786 |
+
"loss": 3.8518,
|
787 |
+
"step": 11000
|
788 |
+
},
|
789 |
+
{
|
790 |
+
"epoch": 1.7703349282296652,
|
791 |
+
"grad_norm": 646417.375,
|
792 |
+
"learning_rate": 3.5247208931419464e-05,
|
793 |
+
"loss": 3.7623,
|
794 |
+
"step": 11100
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 1.7862838915470496,
|
798 |
+
"grad_norm": 641868.25,
|
799 |
+
"learning_rate": 3.511430090377459e-05,
|
800 |
+
"loss": 3.7536,
|
801 |
+
"step": 11200
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 1.802232854864434,
|
805 |
+
"grad_norm": 743804.375,
|
806 |
+
"learning_rate": 3.498139287612972e-05,
|
807 |
+
"loss": 3.8056,
|
808 |
+
"step": 11300
|
809 |
+
},
|
810 |
+
{
|
811 |
+
"epoch": 1.8181818181818183,
|
812 |
+
"grad_norm": 725540.5625,
|
813 |
+
"learning_rate": 3.484848484848485e-05,
|
814 |
+
"loss": 3.8078,
|
815 |
+
"step": 11400
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 1.8341307814992025,
|
819 |
+
"grad_norm": 735264.125,
|
820 |
+
"learning_rate": 3.4715576820839976e-05,
|
821 |
+
"loss": 3.8247,
|
822 |
+
"step": 11500
|
823 |
+
},
|
824 |
+
{
|
825 |
+
"epoch": 1.8500797448165869,
|
826 |
+
"grad_norm": 708785.5625,
|
827 |
+
"learning_rate": 3.458266879319511e-05,
|
828 |
+
"loss": 3.7898,
|
829 |
+
"step": 11600
|
830 |
+
},
|
831 |
+
{
|
832 |
+
"epoch": 1.8660287081339713,
|
833 |
+
"grad_norm": 864416.3125,
|
834 |
+
"learning_rate": 3.444976076555024e-05,
|
835 |
+
"loss": 3.8642,
|
836 |
+
"step": 11700
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 1.8819776714513556,
|
840 |
+
"grad_norm": 693382.5625,
|
841 |
+
"learning_rate": 3.431685273790537e-05,
|
842 |
+
"loss": 3.7433,
|
843 |
+
"step": 11800
|
844 |
+
},
|
845 |
+
{
|
846 |
+
"epoch": 1.89792663476874,
|
847 |
+
"grad_norm": 841755.25,
|
848 |
+
"learning_rate": 3.41839447102605e-05,
|
849 |
+
"loss": 3.8562,
|
850 |
+
"step": 11900
|
851 |
+
},
|
852 |
+
{
|
853 |
+
"epoch": 1.9138755980861244,
|
854 |
+
"grad_norm": 662765.125,
|
855 |
+
"learning_rate": 3.4051036682615634e-05,
|
856 |
+
"loss": 3.8703,
|
857 |
+
"step": 12000
|
858 |
+
},
|
859 |
+
{
|
860 |
+
"epoch": 1.9298245614035088,
|
861 |
+
"grad_norm": 707709.75,
|
862 |
+
"learning_rate": 3.391812865497076e-05,
|
863 |
+
"loss": 3.8402,
|
864 |
+
"step": 12100
|
865 |
+
},
|
866 |
+
{
|
867 |
+
"epoch": 1.9457735247208932,
|
868 |
+
"grad_norm": 616722.875,
|
869 |
+
"learning_rate": 3.3785220627325893e-05,
|
870 |
+
"loss": 3.8432,
|
871 |
+
"step": 12200
|
872 |
+
},
|
873 |
+
{
|
874 |
+
"epoch": 1.9617224880382775,
|
875 |
+
"grad_norm": 795238.625,
|
876 |
+
"learning_rate": 3.3652312599681026e-05,
|
877 |
+
"loss": 3.7319,
|
878 |
+
"step": 12300
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 1.977671451355662,
|
882 |
+
"grad_norm": 709130.0,
|
883 |
+
"learning_rate": 3.351940457203615e-05,
|
884 |
+
"loss": 3.7628,
|
885 |
+
"step": 12400
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 1.9936204146730463,
|
889 |
+
"grad_norm": 789321.375,
|
890 |
+
"learning_rate": 3.3386496544391286e-05,
|
891 |
+
"loss": 3.7821,
|
892 |
+
"step": 12500
|
893 |
+
},
|
894 |
+
{
|
895 |
+
"epoch": 2.0,
|
896 |
+
"eval_loss": 3.800079584121704,
|
897 |
+
"eval_runtime": 212.9573,
|
898 |
+
"eval_samples_per_second": 134.407,
|
899 |
+
"eval_steps_per_second": 4.203,
|
900 |
+
"step": 12540
|
901 |
+
},
|
902 |
+
{
|
903 |
+
"epoch": 2.0095693779904304,
|
904 |
+
"grad_norm": 862786.0,
|
905 |
+
"learning_rate": 3.325358851674641e-05,
|
906 |
+
"loss": 3.7512,
|
907 |
+
"step": 12600
|
908 |
+
},
|
909 |
+
{
|
910 |
+
"epoch": 2.025518341307815,
|
911 |
+
"grad_norm": 772642.5625,
|
912 |
+
"learning_rate": 3.312068048910154e-05,
|
913 |
+
"loss": 3.6611,
|
914 |
+
"step": 12700
|
915 |
+
},
|
916 |
+
{
|
917 |
+
"epoch": 2.041467304625199,
|
918 |
+
"grad_norm": 706907.1875,
|
919 |
+
"learning_rate": 3.298777246145667e-05,
|
920 |
+
"loss": 3.7484,
|
921 |
+
"step": 12800
|
922 |
+
},
|
923 |
+
{
|
924 |
+
"epoch": 2.0574162679425836,
|
925 |
+
"grad_norm": 940896.375,
|
926 |
+
"learning_rate": 3.2854864433811804e-05,
|
927 |
+
"loss": 3.7545,
|
928 |
+
"step": 12900
|
929 |
+
},
|
930 |
+
{
|
931 |
+
"epoch": 2.073365231259968,
|
932 |
+
"grad_norm": 670013.875,
|
933 |
+
"learning_rate": 3.272195640616693e-05,
|
934 |
+
"loss": 3.6504,
|
935 |
+
"step": 13000
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 2.0893141945773523,
|
939 |
+
"grad_norm": 702168.0,
|
940 |
+
"learning_rate": 3.2589048378522064e-05,
|
941 |
+
"loss": 3.6133,
|
942 |
+
"step": 13100
|
943 |
+
},
|
944 |
+
{
|
945 |
+
"epoch": 2.1052631578947367,
|
946 |
+
"grad_norm": 741310.75,
|
947 |
+
"learning_rate": 3.24561403508772e-05,
|
948 |
+
"loss": 3.7202,
|
949 |
+
"step": 13200
|
950 |
+
},
|
951 |
+
{
|
952 |
+
"epoch": 2.121212121212121,
|
953 |
+
"grad_norm": 630812.625,
|
954 |
+
"learning_rate": 3.232323232323233e-05,
|
955 |
+
"loss": 3.7131,
|
956 |
+
"step": 13300
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 2.1371610845295055,
|
960 |
+
"grad_norm": 736768.125,
|
961 |
+
"learning_rate": 3.2190324295587456e-05,
|
962 |
+
"loss": 3.6982,
|
963 |
+
"step": 13400
|
964 |
+
},
|
965 |
+
{
|
966 |
+
"epoch": 2.15311004784689,
|
967 |
+
"grad_norm": 743122.8125,
|
968 |
+
"learning_rate": 3.205741626794259e-05,
|
969 |
+
"loss": 3.7196,
|
970 |
+
"step": 13500
|
971 |
+
},
|
972 |
+
{
|
973 |
+
"epoch": 2.1690590111642742,
|
974 |
+
"grad_norm": 685713.25,
|
975 |
+
"learning_rate": 3.1924508240297715e-05,
|
976 |
+
"loss": 3.7312,
|
977 |
+
"step": 13600
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 2.1850079744816586,
|
981 |
+
"grad_norm": 829695.75,
|
982 |
+
"learning_rate": 3.179160021265284e-05,
|
983 |
+
"loss": 3.6914,
|
984 |
+
"step": 13700
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 2.200956937799043,
|
988 |
+
"grad_norm": 640866.6875,
|
989 |
+
"learning_rate": 3.1658692185007975e-05,
|
990 |
+
"loss": 3.8376,
|
991 |
+
"step": 13800
|
992 |
+
},
|
993 |
+
{
|
994 |
+
"epoch": 2.2169059011164274,
|
995 |
+
"grad_norm": 705478.25,
|
996 |
+
"learning_rate": 3.152578415736311e-05,
|
997 |
+
"loss": 3.7294,
|
998 |
+
"step": 13900
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 2.2328548644338118,
|
1002 |
+
"grad_norm": 664668.8125,
|
1003 |
+
"learning_rate": 3.1392876129718234e-05,
|
1004 |
+
"loss": 3.7046,
|
1005 |
+
"step": 14000
|
1006 |
+
},
|
1007 |
+
{
|
1008 |
+
"epoch": 2.248803827751196,
|
1009 |
+
"grad_norm": 653658.25,
|
1010 |
+
"learning_rate": 3.125996810207337e-05,
|
1011 |
+
"loss": 3.7142,
|
1012 |
+
"step": 14100
|
1013 |
+
},
|
1014 |
+
{
|
1015 |
+
"epoch": 2.2647527910685805,
|
1016 |
+
"grad_norm": 737460.0625,
|
1017 |
+
"learning_rate": 3.11270600744285e-05,
|
1018 |
+
"loss": 3.6449,
|
1019 |
+
"step": 14200
|
1020 |
+
},
|
1021 |
+
{
|
1022 |
+
"epoch": 2.280701754385965,
|
1023 |
+
"grad_norm": 657366.8125,
|
1024 |
+
"learning_rate": 3.0994152046783626e-05,
|
1025 |
+
"loss": 3.7051,
|
1026 |
+
"step": 14300
|
1027 |
+
},
|
1028 |
+
{
|
1029 |
+
"epoch": 2.2966507177033493,
|
1030 |
+
"grad_norm": 734922.0625,
|
1031 |
+
"learning_rate": 3.086124401913876e-05,
|
1032 |
+
"loss": 3.6793,
|
1033 |
+
"step": 14400
|
1034 |
+
},
|
1035 |
+
{
|
1036 |
+
"epoch": 2.3125996810207337,
|
1037 |
+
"grad_norm": 636888.125,
|
1038 |
+
"learning_rate": 3.072833599149389e-05,
|
1039 |
+
"loss": 3.764,
|
1040 |
+
"step": 14500
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 2.328548644338118,
|
1044 |
+
"grad_norm": 776567.125,
|
1045 |
+
"learning_rate": 3.059542796384902e-05,
|
1046 |
+
"loss": 3.6952,
|
1047 |
+
"step": 14600
|
1048 |
+
},
|
1049 |
+
{
|
1050 |
+
"epoch": 2.3444976076555024,
|
1051 |
+
"grad_norm": 678031.5,
|
1052 |
+
"learning_rate": 3.0462519936204148e-05,
|
1053 |
+
"loss": 3.7086,
|
1054 |
+
"step": 14700
|
1055 |
+
},
|
1056 |
+
{
|
1057 |
+
"epoch": 2.360446570972887,
|
1058 |
+
"grad_norm": 702883.9375,
|
1059 |
+
"learning_rate": 3.032961190855928e-05,
|
1060 |
+
"loss": 3.6974,
|
1061 |
+
"step": 14800
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 2.376395534290271,
|
1065 |
+
"grad_norm": 640335.375,
|
1066 |
+
"learning_rate": 3.0196703880914407e-05,
|
1067 |
+
"loss": 3.6555,
|
1068 |
+
"step": 14900
|
1069 |
+
},
|
1070 |
+
{
|
1071 |
+
"epoch": 2.3923444976076556,
|
1072 |
+
"grad_norm": 686070.0625,
|
1073 |
+
"learning_rate": 3.0063795853269537e-05,
|
1074 |
+
"loss": 3.7167,
|
1075 |
+
"step": 15000
|
1076 |
+
},
|
1077 |
+
{
|
1078 |
+
"epoch": 2.40829346092504,
|
1079 |
+
"grad_norm": 676044.25,
|
1080 |
+
"learning_rate": 2.993088782562467e-05,
|
1081 |
+
"loss": 3.6554,
|
1082 |
+
"step": 15100
|
1083 |
+
},
|
1084 |
+
{
|
1085 |
+
"epoch": 2.4242424242424243,
|
1086 |
+
"grad_norm": 767818.6875,
|
1087 |
+
"learning_rate": 2.9797979797979796e-05,
|
1088 |
+
"loss": 3.693,
|
1089 |
+
"step": 15200
|
1090 |
+
},
|
1091 |
+
{
|
1092 |
+
"epoch": 2.4401913875598087,
|
1093 |
+
"grad_norm": 739911.0625,
|
1094 |
+
"learning_rate": 2.966507177033493e-05,
|
1095 |
+
"loss": 3.6431,
|
1096 |
+
"step": 15300
|
1097 |
+
},
|
1098 |
+
{
|
1099 |
+
"epoch": 2.456140350877193,
|
1100 |
+
"grad_norm": 771256.3125,
|
1101 |
+
"learning_rate": 2.9532163742690062e-05,
|
1102 |
+
"loss": 3.6216,
|
1103 |
+
"step": 15400
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 2.4720893141945774,
|
1107 |
+
"grad_norm": 773173.375,
|
1108 |
+
"learning_rate": 2.939925571504519e-05,
|
1109 |
+
"loss": 3.7742,
|
1110 |
+
"step": 15500
|
1111 |
+
},
|
1112 |
+
{
|
1113 |
+
"epoch": 2.488038277511962,
|
1114 |
+
"grad_norm": 762888.3125,
|
1115 |
+
"learning_rate": 2.9266347687400318e-05,
|
1116 |
+
"loss": 3.7236,
|
1117 |
+
"step": 15600
|
1118 |
+
},
|
1119 |
+
{
|
1120 |
+
"epoch": 2.503987240829346,
|
1121 |
+
"grad_norm": 879857.0625,
|
1122 |
+
"learning_rate": 2.913343965975545e-05,
|
1123 |
+
"loss": 3.7253,
|
1124 |
+
"step": 15700
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 2.5199362041467306,
|
1128 |
+
"grad_norm": 733285.0625,
|
1129 |
+
"learning_rate": 2.9000531632110584e-05,
|
1130 |
+
"loss": 3.7166,
|
1131 |
+
"step": 15800
|
1132 |
+
},
|
1133 |
+
{
|
1134 |
+
"epoch": 2.535885167464115,
|
1135 |
+
"grad_norm": 833014.3125,
|
1136 |
+
"learning_rate": 2.886762360446571e-05,
|
1137 |
+
"loss": 3.6705,
|
1138 |
+
"step": 15900
|
1139 |
+
},
|
1140 |
+
{
|
1141 |
+
"epoch": 2.5518341307814993,
|
1142 |
+
"grad_norm": 736943.4375,
|
1143 |
+
"learning_rate": 2.8734715576820844e-05,
|
1144 |
+
"loss": 3.7253,
|
1145 |
+
"step": 16000
|
1146 |
+
},
|
1147 |
+
{
|
1148 |
+
"epoch": 2.5677830940988837,
|
1149 |
+
"grad_norm": 708644.4375,
|
1150 |
+
"learning_rate": 2.8601807549175973e-05,
|
1151 |
+
"loss": 3.6672,
|
1152 |
+
"step": 16100
|
1153 |
+
},
|
1154 |
+
{
|
1155 |
+
"epoch": 2.583732057416268,
|
1156 |
+
"grad_norm": 795735.8125,
|
1157 |
+
"learning_rate": 2.84688995215311e-05,
|
1158 |
+
"loss": 3.728,
|
1159 |
+
"step": 16200
|
1160 |
+
},
|
1161 |
+
{
|
1162 |
+
"epoch": 2.5996810207336525,
|
1163 |
+
"grad_norm": 864533.875,
|
1164 |
+
"learning_rate": 2.8335991493886233e-05,
|
1165 |
+
"loss": 3.6478,
|
1166 |
+
"step": 16300
|
1167 |
+
},
|
1168 |
+
{
|
1169 |
+
"epoch": 2.6156299840510364,
|
1170 |
+
"grad_norm": 835745.0,
|
1171 |
+
"learning_rate": 2.8203083466241366e-05,
|
1172 |
+
"loss": 3.6204,
|
1173 |
+
"step": 16400
|
1174 |
+
},
|
1175 |
+
{
|
1176 |
+
"epoch": 2.6315789473684212,
|
1177 |
+
"grad_norm": 726415.3125,
|
1178 |
+
"learning_rate": 2.8070175438596492e-05,
|
1179 |
+
"loss": 3.6899,
|
1180 |
+
"step": 16500
|
1181 |
+
},
|
1182 |
+
{
|
1183 |
+
"epoch": 2.647527910685805,
|
1184 |
+
"grad_norm": 619188.1875,
|
1185 |
+
"learning_rate": 2.793726741095162e-05,
|
1186 |
+
"loss": 3.6442,
|
1187 |
+
"step": 16600
|
1188 |
+
},
|
1189 |
+
{
|
1190 |
+
"epoch": 2.66347687400319,
|
1191 |
+
"grad_norm": 704278.1875,
|
1192 |
+
"learning_rate": 2.7804359383306755e-05,
|
1193 |
+
"loss": 3.7113,
|
1194 |
+
"step": 16700
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 2.679425837320574,
|
1198 |
+
"grad_norm": 787753.0,
|
1199 |
+
"learning_rate": 2.767145135566188e-05,
|
1200 |
+
"loss": 3.7368,
|
1201 |
+
"step": 16800
|
1202 |
+
},
|
1203 |
+
{
|
1204 |
+
"epoch": 2.6953748006379588,
|
1205 |
+
"grad_norm": 738743.5625,
|
1206 |
+
"learning_rate": 2.7538543328017014e-05,
|
1207 |
+
"loss": 3.7575,
|
1208 |
+
"step": 16900
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 2.7113237639553427,
|
1212 |
+
"grad_norm": 797812.75,
|
1213 |
+
"learning_rate": 2.7405635300372147e-05,
|
1214 |
+
"loss": 3.6586,
|
1215 |
+
"step": 17000
|
1216 |
+
},
|
1217 |
+
{
|
1218 |
+
"epoch": 2.7272727272727275,
|
1219 |
+
"grad_norm": 662883.0,
|
1220 |
+
"learning_rate": 2.7272727272727273e-05,
|
1221 |
+
"loss": 3.6812,
|
1222 |
+
"step": 17100
|
1223 |
+
},
|
1224 |
+
{
|
1225 |
+
"epoch": 2.7432216905901115,
|
1226 |
+
"grad_norm": 708693.8125,
|
1227 |
+
"learning_rate": 2.7139819245082403e-05,
|
1228 |
+
"loss": 3.6645,
|
1229 |
+
"step": 17200
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 2.7591706539074963,
|
1233 |
+
"grad_norm": 750630.625,
|
1234 |
+
"learning_rate": 2.7006911217437536e-05,
|
1235 |
+
"loss": 3.659,
|
1236 |
+
"step": 17300
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 2.77511961722488,
|
1240 |
+
"grad_norm": 743643.5625,
|
1241 |
+
"learning_rate": 2.6874003189792662e-05,
|
1242 |
+
"loss": 3.6794,
|
1243 |
+
"step": 17400
|
1244 |
+
},
|
1245 |
+
{
|
1246 |
+
"epoch": 2.7910685805422646,
|
1247 |
+
"grad_norm": 683095.9375,
|
1248 |
+
"learning_rate": 2.6741095162147795e-05,
|
1249 |
+
"loss": 3.6587,
|
1250 |
+
"step": 17500
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 2.807017543859649,
|
1254 |
+
"grad_norm": 659289.8125,
|
1255 |
+
"learning_rate": 2.6608187134502928e-05,
|
1256 |
+
"loss": 3.7067,
|
1257 |
+
"step": 17600
|
1258 |
+
},
|
1259 |
+
{
|
1260 |
+
"epoch": 2.8229665071770333,
|
1261 |
+
"grad_norm": 722875.9375,
|
1262 |
+
"learning_rate": 2.6475279106858054e-05,
|
1263 |
+
"loss": 3.6918,
|
1264 |
+
"step": 17700
|
1265 |
+
},
|
1266 |
+
{
|
1267 |
+
"epoch": 2.8389154704944177,
|
1268 |
+
"grad_norm": 643060.4375,
|
1269 |
+
"learning_rate": 2.6342371079213184e-05,
|
1270 |
+
"loss": 3.694,
|
1271 |
+
"step": 17800
|
1272 |
+
},
|
1273 |
+
{
|
1274 |
+
"epoch": 2.854864433811802,
|
1275 |
+
"grad_norm": 811117.3125,
|
1276 |
+
"learning_rate": 2.6209463051568317e-05,
|
1277 |
+
"loss": 3.703,
|
1278 |
+
"step": 17900
|
1279 |
+
},
|
1280 |
+
{
|
1281 |
+
"epoch": 2.8708133971291865,
|
1282 |
+
"grad_norm": 835076.6875,
|
1283 |
+
"learning_rate": 2.6076555023923443e-05,
|
1284 |
+
"loss": 3.6346,
|
1285 |
+
"step": 18000
|
1286 |
+
},
|
1287 |
+
{
|
1288 |
+
"epoch": 2.886762360446571,
|
1289 |
+
"grad_norm": 800111.875,
|
1290 |
+
"learning_rate": 2.5943646996278576e-05,
|
1291 |
+
"loss": 3.6589,
|
1292 |
+
"step": 18100
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 2.9027113237639552,
|
1296 |
+
"grad_norm": 726367.5,
|
1297 |
+
"learning_rate": 2.5810738968633706e-05,
|
1298 |
+
"loss": 3.6522,
|
1299 |
+
"step": 18200
|
1300 |
+
},
|
1301 |
+
{
|
1302 |
+
"epoch": 2.9186602870813396,
|
1303 |
+
"grad_norm": 717922.25,
|
1304 |
+
"learning_rate": 2.567783094098884e-05,
|
1305 |
+
"loss": 3.6592,
|
1306 |
+
"step": 18300
|
1307 |
+
},
|
1308 |
+
{
|
1309 |
+
"epoch": 2.934609250398724,
|
1310 |
+
"grad_norm": 841302.0625,
|
1311 |
+
"learning_rate": 2.5544922913343965e-05,
|
1312 |
+
"loss": 3.688,
|
1313 |
+
"step": 18400
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 2.9505582137161084,
|
1317 |
+
"grad_norm": 697304.1875,
|
1318 |
+
"learning_rate": 2.5412014885699098e-05,
|
1319 |
+
"loss": 3.6017,
|
1320 |
+
"step": 18500
|
1321 |
+
},
|
1322 |
+
{
|
1323 |
+
"epoch": 2.9665071770334928,
|
1324 |
+
"grad_norm": 929050.25,
|
1325 |
+
"learning_rate": 2.527910685805423e-05,
|
1326 |
+
"loss": 3.6116,
|
1327 |
+
"step": 18600
|
1328 |
+
},
|
1329 |
+
{
|
1330 |
+
"epoch": 2.982456140350877,
|
1331 |
+
"grad_norm": 699048.9375,
|
1332 |
+
"learning_rate": 2.5146198830409358e-05,
|
1333 |
+
"loss": 3.6474,
|
1334 |
+
"step": 18700
|
1335 |
+
},
|
1336 |
+
{
|
1337 |
+
"epoch": 2.9984051036682615,
|
1338 |
+
"grad_norm": 783686.0625,
|
1339 |
+
"learning_rate": 2.5013290802764487e-05,
|
1340 |
+
"loss": 3.6783,
|
1341 |
+
"step": 18800
|
1342 |
+
},
|
1343 |
+
{
|
1344 |
+
"epoch": 3.0,
|
1345 |
+
"eval_loss": 3.706315755844116,
|
1346 |
+
"eval_runtime": 212.8533,
|
1347 |
+
"eval_samples_per_second": 134.473,
|
1348 |
+
"eval_steps_per_second": 4.205,
|
1349 |
+
"step": 18810
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 3.014354066985646,
|
1353 |
+
"grad_norm": 709059.3125,
|
1354 |
+
"learning_rate": 2.4880382775119617e-05,
|
1355 |
+
"loss": 3.5423,
|
1356 |
+
"step": 18900
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 3.0303030303030303,
|
1360 |
+
"grad_norm": 655684.75,
|
1361 |
+
"learning_rate": 2.474747474747475e-05,
|
1362 |
+
"loss": 3.5606,
|
1363 |
+
"step": 19000
|
1364 |
+
},
|
1365 |
+
{
|
1366 |
+
"epoch": 3.0462519936204147,
|
1367 |
+
"grad_norm": 759915.1875,
|
1368 |
+
"learning_rate": 2.461456671982988e-05,
|
1369 |
+
"loss": 3.6979,
|
1370 |
+
"step": 19100
|
1371 |
+
},
|
1372 |
+
{
|
1373 |
+
"epoch": 3.062200956937799,
|
1374 |
+
"grad_norm": 835860.375,
|
1375 |
+
"learning_rate": 2.448165869218501e-05,
|
1376 |
+
"loss": 3.6324,
|
1377 |
+
"step": 19200
|
1378 |
+
},
|
1379 |
+
{
|
1380 |
+
"epoch": 3.0781499202551834,
|
1381 |
+
"grad_norm": 675779.5625,
|
1382 |
+
"learning_rate": 2.434875066454014e-05,
|
1383 |
+
"loss": 3.5773,
|
1384 |
+
"step": 19300
|
1385 |
+
},
|
1386 |
+
{
|
1387 |
+
"epoch": 3.094098883572568,
|
1388 |
+
"grad_norm": 700443.875,
|
1389 |
+
"learning_rate": 2.421584263689527e-05,
|
1390 |
+
"loss": 3.6309,
|
1391 |
+
"step": 19400
|
1392 |
+
},
|
1393 |
+
{
|
1394 |
+
"epoch": 3.110047846889952,
|
1395 |
+
"grad_norm": 768801.6875,
|
1396 |
+
"learning_rate": 2.4082934609250398e-05,
|
1397 |
+
"loss": 3.639,
|
1398 |
+
"step": 19500
|
1399 |
+
},
|
1400 |
+
{
|
1401 |
+
"epoch": 3.1259968102073366,
|
1402 |
+
"grad_norm": 705109.6875,
|
1403 |
+
"learning_rate": 2.395002658160553e-05,
|
1404 |
+
"loss": 3.6365,
|
1405 |
+
"step": 19600
|
1406 |
+
},
|
1407 |
+
{
|
1408 |
+
"epoch": 3.141945773524721,
|
1409 |
+
"grad_norm": 661887.5,
|
1410 |
+
"learning_rate": 2.381711855396066e-05,
|
1411 |
+
"loss": 3.5754,
|
1412 |
+
"step": 19700
|
1413 |
+
},
|
1414 |
+
{
|
1415 |
+
"epoch": 3.1578947368421053,
|
1416 |
+
"grad_norm": 691131.0,
|
1417 |
+
"learning_rate": 2.368421052631579e-05,
|
1418 |
+
"loss": 3.5888,
|
1419 |
+
"step": 19800
|
1420 |
+
},
|
1421 |
+
{
|
1422 |
+
"epoch": 3.1738437001594897,
|
1423 |
+
"grad_norm": 738841.625,
|
1424 |
+
"learning_rate": 2.355130249867092e-05,
|
1425 |
+
"loss": 3.6433,
|
1426 |
+
"step": 19900
|
1427 |
+
},
|
1428 |
+
{
|
1429 |
+
"epoch": 3.189792663476874,
|
1430 |
+
"grad_norm": 663372.1875,
|
1431 |
+
"learning_rate": 2.341839447102605e-05,
|
1432 |
+
"loss": 3.5642,
|
1433 |
+
"step": 20000
|
1434 |
+
},
|
1435 |
+
{
|
1436 |
+
"epoch": 3.2057416267942584,
|
1437 |
+
"grad_norm": 619548.75,
|
1438 |
+
"learning_rate": 2.3285486443381183e-05,
|
1439 |
+
"loss": 3.5864,
|
1440 |
+
"step": 20100
|
1441 |
+
},
|
1442 |
+
{
|
1443 |
+
"epoch": 3.221690590111643,
|
1444 |
+
"grad_norm": 773919.6875,
|
1445 |
+
"learning_rate": 2.3152578415736312e-05,
|
1446 |
+
"loss": 3.6072,
|
1447 |
+
"step": 20200
|
1448 |
+
},
|
1449 |
+
{
|
1450 |
+
"epoch": 3.237639553429027,
|
1451 |
+
"grad_norm": 728946.0625,
|
1452 |
+
"learning_rate": 2.3019670388091442e-05,
|
1453 |
+
"loss": 3.6212,
|
1454 |
+
"step": 20300
|
1455 |
+
},
|
1456 |
+
{
|
1457 |
+
"epoch": 3.2535885167464116,
|
1458 |
+
"grad_norm": 659326.9375,
|
1459 |
+
"learning_rate": 2.288676236044657e-05,
|
1460 |
+
"loss": 3.6068,
|
1461 |
+
"step": 20400
|
1462 |
+
},
|
1463 |
+
{
|
1464 |
+
"epoch": 3.269537480063796,
|
1465 |
+
"grad_norm": 759802.25,
|
1466 |
+
"learning_rate": 2.27538543328017e-05,
|
1467 |
+
"loss": 3.6783,
|
1468 |
+
"step": 20500
|
1469 |
+
},
|
1470 |
+
{
|
1471 |
+
"epoch": 3.2854864433811803,
|
1472 |
+
"grad_norm": 805141.9375,
|
1473 |
+
"learning_rate": 2.262094630515683e-05,
|
1474 |
+
"loss": 3.6441,
|
1475 |
+
"step": 20600
|
1476 |
+
},
|
1477 |
+
{
|
1478 |
+
"epoch": 3.3014354066985647,
|
1479 |
+
"grad_norm": 716521.875,
|
1480 |
+
"learning_rate": 2.2488038277511964e-05,
|
1481 |
+
"loss": 3.5748,
|
1482 |
+
"step": 20700
|
1483 |
+
},
|
1484 |
+
{
|
1485 |
+
"epoch": 3.317384370015949,
|
1486 |
+
"grad_norm": 920115.4375,
|
1487 |
+
"learning_rate": 2.2355130249867094e-05,
|
1488 |
+
"loss": 3.5985,
|
1489 |
+
"step": 20800
|
1490 |
+
},
|
1491 |
+
{
|
1492 |
+
"epoch": 3.3333333333333335,
|
1493 |
+
"grad_norm": 838658.0625,
|
1494 |
+
"learning_rate": 2.2222222222222223e-05,
|
1495 |
+
"loss": 3.564,
|
1496 |
+
"step": 20900
|
1497 |
+
},
|
1498 |
+
{
|
1499 |
+
"epoch": 3.349282296650718,
|
1500 |
+
"grad_norm": 838474.9375,
|
1501 |
+
"learning_rate": 2.2089314194577353e-05,
|
1502 |
+
"loss": 3.5875,
|
1503 |
+
"step": 21000
|
1504 |
+
},
|
1505 |
+
{
|
1506 |
+
"epoch": 3.3652312599681022,
|
1507 |
+
"grad_norm": 663992.25,
|
1508 |
+
"learning_rate": 2.1956406166932483e-05,
|
1509 |
+
"loss": 3.5681,
|
1510 |
+
"step": 21100
|
1511 |
+
},
|
1512 |
+
{
|
1513 |
+
"epoch": 3.3811802232854866,
|
1514 |
+
"grad_norm": 679489.9375,
|
1515 |
+
"learning_rate": 2.1823498139287616e-05,
|
1516 |
+
"loss": 3.5276,
|
1517 |
+
"step": 21200
|
1518 |
+
},
|
1519 |
+
{
|
1520 |
+
"epoch": 3.397129186602871,
|
1521 |
+
"grad_norm": 721495.6875,
|
1522 |
+
"learning_rate": 2.1690590111642745e-05,
|
1523 |
+
"loss": 3.5599,
|
1524 |
+
"step": 21300
|
1525 |
+
},
|
1526 |
+
{
|
1527 |
+
"epoch": 3.4130781499202554,
|
1528 |
+
"grad_norm": 667400.9375,
|
1529 |
+
"learning_rate": 2.1557682083997875e-05,
|
1530 |
+
"loss": 3.5554,
|
1531 |
+
"step": 21400
|
1532 |
+
},
|
1533 |
+
{
|
1534 |
+
"epoch": 3.4290271132376393,
|
1535 |
+
"grad_norm": 706122.0,
|
1536 |
+
"learning_rate": 2.1424774056353005e-05,
|
1537 |
+
"loss": 3.6543,
|
1538 |
+
"step": 21500
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 3.444976076555024,
|
1542 |
+
"grad_norm": 709513.125,
|
1543 |
+
"learning_rate": 2.1291866028708134e-05,
|
1544 |
+
"loss": 3.6568,
|
1545 |
+
"step": 21600
|
1546 |
+
},
|
1547 |
+
{
|
1548 |
+
"epoch": 3.460925039872408,
|
1549 |
+
"grad_norm": 652669.375,
|
1550 |
+
"learning_rate": 2.1158958001063264e-05,
|
1551 |
+
"loss": 3.6423,
|
1552 |
+
"step": 21700
|
1553 |
+
},
|
1554 |
+
{
|
1555 |
+
"epoch": 3.476874003189793,
|
1556 |
+
"grad_norm": 713371.625,
|
1557 |
+
"learning_rate": 2.1026049973418397e-05,
|
1558 |
+
"loss": 3.6102,
|
1559 |
+
"step": 21800
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 3.492822966507177,
|
1563 |
+
"grad_norm": 694536.875,
|
1564 |
+
"learning_rate": 2.0893141945773527e-05,
|
1565 |
+
"loss": 3.5488,
|
1566 |
+
"step": 21900
|
1567 |
+
},
|
1568 |
+
{
|
1569 |
+
"epoch": 3.5087719298245617,
|
1570 |
+
"grad_norm": 668059.9375,
|
1571 |
+
"learning_rate": 2.0760233918128656e-05,
|
1572 |
+
"loss": 3.6078,
|
1573 |
+
"step": 22000
|
1574 |
+
},
|
1575 |
+
{
|
1576 |
+
"epoch": 3.5247208931419456,
|
1577 |
+
"grad_norm": 731925.875,
|
1578 |
+
"learning_rate": 2.0627325890483786e-05,
|
1579 |
+
"loss": 3.5455,
|
1580 |
+
"step": 22100
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 3.5406698564593304,
|
1584 |
+
"grad_norm": 766777.9375,
|
1585 |
+
"learning_rate": 2.0494417862838915e-05,
|
1586 |
+
"loss": 3.5582,
|
1587 |
+
"step": 22200
|
1588 |
+
},
|
1589 |
+
{
|
1590 |
+
"epoch": 3.5566188197767143,
|
1591 |
+
"grad_norm": 755381.1875,
|
1592 |
+
"learning_rate": 2.0361509835194045e-05,
|
1593 |
+
"loss": 3.5859,
|
1594 |
+
"step": 22300
|
1595 |
+
},
|
1596 |
+
{
|
1597 |
+
"epoch": 3.5725677830940987,
|
1598 |
+
"grad_norm": 749873.6875,
|
1599 |
+
"learning_rate": 2.0228601807549178e-05,
|
1600 |
+
"loss": 3.7161,
|
1601 |
+
"step": 22400
|
1602 |
+
},
|
1603 |
+
{
|
1604 |
+
"epoch": 3.588516746411483,
|
1605 |
+
"grad_norm": 677539.125,
|
1606 |
+
"learning_rate": 2.0095693779904308e-05,
|
1607 |
+
"loss": 3.6225,
|
1608 |
+
"step": 22500
|
1609 |
+
},
|
1610 |
+
{
|
1611 |
+
"epoch": 3.6044657097288675,
|
1612 |
+
"grad_norm": 744990.8125,
|
1613 |
+
"learning_rate": 1.9962785752259437e-05,
|
1614 |
+
"loss": 3.5935,
|
1615 |
+
"step": 22600
|
1616 |
+
},
|
1617 |
+
{
|
1618 |
+
"epoch": 3.620414673046252,
|
1619 |
+
"grad_norm": 629363.0,
|
1620 |
+
"learning_rate": 1.9829877724614567e-05,
|
1621 |
+
"loss": 3.5828,
|
1622 |
+
"step": 22700
|
1623 |
+
},
|
1624 |
+
{
|
1625 |
+
"epoch": 3.6363636363636362,
|
1626 |
+
"grad_norm": 685327.4375,
|
1627 |
+
"learning_rate": 1.9696969696969697e-05,
|
1628 |
+
"loss": 3.5937,
|
1629 |
+
"step": 22800
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 3.6523125996810206,
|
1633 |
+
"grad_norm": 760474.25,
|
1634 |
+
"learning_rate": 1.956406166932483e-05,
|
1635 |
+
"loss": 3.5799,
|
1636 |
+
"step": 22900
|
1637 |
+
},
|
1638 |
+
{
|
1639 |
+
"epoch": 3.668261562998405,
|
1640 |
+
"grad_norm": 725888.9375,
|
1641 |
+
"learning_rate": 1.943115364167996e-05,
|
1642 |
+
"loss": 3.5465,
|
1643 |
+
"step": 23000
|
1644 |
+
},
|
1645 |
+
{
|
1646 |
+
"epoch": 3.6842105263157894,
|
1647 |
+
"grad_norm": 752312.0625,
|
1648 |
+
"learning_rate": 1.929824561403509e-05,
|
1649 |
+
"loss": 3.6185,
|
1650 |
+
"step": 23100
|
1651 |
+
},
|
1652 |
+
{
|
1653 |
+
"epoch": 3.7001594896331738,
|
1654 |
+
"grad_norm": 855498.5625,
|
1655 |
+
"learning_rate": 1.916533758639022e-05,
|
1656 |
+
"loss": 3.5792,
|
1657 |
+
"step": 23200
|
1658 |
+
},
|
1659 |
+
{
|
1660 |
+
"epoch": 3.716108452950558,
|
1661 |
+
"grad_norm": 679264.5,
|
1662 |
+
"learning_rate": 1.9032429558745348e-05,
|
1663 |
+
"loss": 3.5531,
|
1664 |
+
"step": 23300
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 3.7320574162679425,
|
1668 |
+
"grad_norm": 688482.875,
|
1669 |
+
"learning_rate": 1.8899521531100478e-05,
|
1670 |
+
"loss": 3.5495,
|
1671 |
+
"step": 23400
|
1672 |
+
},
|
1673 |
+
{
|
1674 |
+
"epoch": 3.748006379585327,
|
1675 |
+
"grad_norm": 756693.75,
|
1676 |
+
"learning_rate": 1.876661350345561e-05,
|
1677 |
+
"loss": 3.567,
|
1678 |
+
"step": 23500
|
1679 |
+
},
|
1680 |
+
{
|
1681 |
+
"epoch": 3.7639553429027113,
|
1682 |
+
"grad_norm": 854033.0625,
|
1683 |
+
"learning_rate": 1.863370547581074e-05,
|
1684 |
+
"loss": 3.5611,
|
1685 |
+
"step": 23600
|
1686 |
+
},
|
1687 |
+
{
|
1688 |
+
"epoch": 3.7799043062200957,
|
1689 |
+
"grad_norm": 753585.0625,
|
1690 |
+
"learning_rate": 1.850079744816587e-05,
|
1691 |
+
"loss": 3.5439,
|
1692 |
+
"step": 23700
|
1693 |
+
},
|
1694 |
+
{
|
1695 |
+
"epoch": 3.79585326953748,
|
1696 |
+
"grad_norm": 685358.25,
|
1697 |
+
"learning_rate": 1.8367889420521e-05,
|
1698 |
+
"loss": 3.535,
|
1699 |
+
"step": 23800
|
1700 |
+
},
|
1701 |
+
{
|
1702 |
+
"epoch": 3.8118022328548644,
|
1703 |
+
"grad_norm": 813028.4375,
|
1704 |
+
"learning_rate": 1.823498139287613e-05,
|
1705 |
+
"loss": 3.566,
|
1706 |
+
"step": 23900
|
1707 |
+
},
|
1708 |
+
{
|
1709 |
+
"epoch": 3.827751196172249,
|
1710 |
+
"grad_norm": 676295.9375,
|
1711 |
+
"learning_rate": 1.8102073365231263e-05,
|
1712 |
+
"loss": 3.5208,
|
1713 |
+
"step": 24000
|
1714 |
+
},
|
1715 |
+
{
|
1716 |
+
"epoch": 3.843700159489633,
|
1717 |
+
"grad_norm": 705614.1875,
|
1718 |
+
"learning_rate": 1.7969165337586392e-05,
|
1719 |
+
"loss": 3.5108,
|
1720 |
+
"step": 24100
|
1721 |
+
},
|
1722 |
+
{
|
1723 |
+
"epoch": 3.8596491228070176,
|
1724 |
+
"grad_norm": 643356.125,
|
1725 |
+
"learning_rate": 1.7836257309941522e-05,
|
1726 |
+
"loss": 3.5416,
|
1727 |
+
"step": 24200
|
1728 |
+
},
|
1729 |
+
{
|
1730 |
+
"epoch": 3.875598086124402,
|
1731 |
+
"grad_norm": 783635.625,
|
1732 |
+
"learning_rate": 1.770334928229665e-05,
|
1733 |
+
"loss": 3.6022,
|
1734 |
+
"step": 24300
|
1735 |
+
},
|
1736 |
+
{
|
1737 |
+
"epoch": 3.8915470494417863,
|
1738 |
+
"grad_norm": 829787.5,
|
1739 |
+
"learning_rate": 1.757044125465178e-05,
|
1740 |
+
"loss": 3.6056,
|
1741 |
+
"step": 24400
|
1742 |
+
},
|
1743 |
+
{
|
1744 |
+
"epoch": 3.9074960127591707,
|
1745 |
+
"grad_norm": 776633.0625,
|
1746 |
+
"learning_rate": 1.743753322700691e-05,
|
1747 |
+
"loss": 3.5857,
|
1748 |
+
"step": 24500
|
1749 |
+
},
|
1750 |
+
{
|
1751 |
+
"epoch": 3.923444976076555,
|
1752 |
+
"grad_norm": 708855.1875,
|
1753 |
+
"learning_rate": 1.7304625199362044e-05,
|
1754 |
+
"loss": 3.6086,
|
1755 |
+
"step": 24600
|
1756 |
+
},
|
1757 |
+
{
|
1758 |
+
"epoch": 3.9393939393939394,
|
1759 |
+
"grad_norm": 906587.0,
|
1760 |
+
"learning_rate": 1.7171717171717173e-05,
|
1761 |
+
"loss": 3.5361,
|
1762 |
+
"step": 24700
|
1763 |
+
},
|
1764 |
+
{
|
1765 |
+
"epoch": 3.955342902711324,
|
1766 |
+
"grad_norm": 835950.4375,
|
1767 |
+
"learning_rate": 1.7038809144072303e-05,
|
1768 |
+
"loss": 3.5976,
|
1769 |
+
"step": 24800
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 3.971291866028708,
|
1773 |
+
"grad_norm": 816570.6875,
|
1774 |
+
"learning_rate": 1.6905901116427433e-05,
|
1775 |
+
"loss": 3.596,
|
1776 |
+
"step": 24900
|
1777 |
+
},
|
1778 |
+
{
|
1779 |
+
"epoch": 3.9872408293460926,
|
1780 |
+
"grad_norm": 751378.9375,
|
1781 |
+
"learning_rate": 1.6772993088782562e-05,
|
1782 |
+
"loss": 3.4856,
|
1783 |
+
"step": 25000
|
1784 |
+
},
|
1785 |
+
{
|
1786 |
+
"epoch": 4.0,
|
1787 |
+
"eval_loss": 3.646216869354248,
|
1788 |
+
"eval_runtime": 212.927,
|
1789 |
+
"eval_samples_per_second": 134.426,
|
1790 |
+
"eval_steps_per_second": 4.203,
|
1791 |
+
"step": 25080
|
1792 |
+
},
|
1793 |
+
{
|
1794 |
+
"epoch": 4.003189792663477,
|
1795 |
+
"grad_norm": 735559.375,
|
1796 |
+
"learning_rate": 1.6640085061137695e-05,
|
1797 |
+
"loss": 3.5286,
|
1798 |
+
"step": 25100
|
1799 |
+
},
|
1800 |
+
{
|
1801 |
+
"epoch": 4.019138755980861,
|
1802 |
+
"grad_norm": 741852.3125,
|
1803 |
+
"learning_rate": 1.6507177033492825e-05,
|
1804 |
+
"loss": 3.5367,
|
1805 |
+
"step": 25200
|
1806 |
+
},
|
1807 |
+
{
|
1808 |
+
"epoch": 4.035087719298246,
|
1809 |
+
"grad_norm": 650216.625,
|
1810 |
+
"learning_rate": 1.6374269005847955e-05,
|
1811 |
+
"loss": 3.5056,
|
1812 |
+
"step": 25300
|
1813 |
+
},
|
1814 |
+
{
|
1815 |
+
"epoch": 4.05103668261563,
|
1816 |
+
"grad_norm": 717938.9375,
|
1817 |
+
"learning_rate": 1.6241360978203084e-05,
|
1818 |
+
"loss": 3.4996,
|
1819 |
+
"step": 25400
|
1820 |
+
},
|
1821 |
+
{
|
1822 |
+
"epoch": 4.0669856459330145,
|
1823 |
+
"grad_norm": 840426.9375,
|
1824 |
+
"learning_rate": 1.6108452950558214e-05,
|
1825 |
+
"loss": 3.5293,
|
1826 |
+
"step": 25500
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 4.082934609250398,
|
1830 |
+
"grad_norm": 699668.5,
|
1831 |
+
"learning_rate": 1.5975544922913344e-05,
|
1832 |
+
"loss": 3.5122,
|
1833 |
+
"step": 25600
|
1834 |
+
},
|
1835 |
+
{
|
1836 |
+
"epoch": 4.098883572567783,
|
1837 |
+
"grad_norm": 649238.8125,
|
1838 |
+
"learning_rate": 1.5842636895268477e-05,
|
1839 |
+
"loss": 3.4949,
|
1840 |
+
"step": 25700
|
1841 |
+
},
|
1842 |
+
{
|
1843 |
+
"epoch": 4.114832535885167,
|
1844 |
+
"grad_norm": 669479.3125,
|
1845 |
+
"learning_rate": 1.5709728867623606e-05,
|
1846 |
+
"loss": 3.5515,
|
1847 |
+
"step": 25800
|
1848 |
+
},
|
1849 |
+
{
|
1850 |
+
"epoch": 4.130781499202552,
|
1851 |
+
"grad_norm": 645182.625,
|
1852 |
+
"learning_rate": 1.5576820839978733e-05,
|
1853 |
+
"loss": 3.5113,
|
1854 |
+
"step": 25900
|
1855 |
+
},
|
1856 |
+
{
|
1857 |
+
"epoch": 4.146730462519936,
|
1858 |
+
"grad_norm": 913136.6875,
|
1859 |
+
"learning_rate": 1.5443912812333866e-05,
|
1860 |
+
"loss": 3.5326,
|
1861 |
+
"step": 26000
|
1862 |
+
},
|
1863 |
+
{
|
1864 |
+
"epoch": 4.162679425837321,
|
1865 |
+
"grad_norm": 641233.3125,
|
1866 |
+
"learning_rate": 1.5311004784688995e-05,
|
1867 |
+
"loss": 3.4768,
|
1868 |
+
"step": 26100
|
1869 |
+
},
|
1870 |
+
{
|
1871 |
+
"epoch": 4.178628389154705,
|
1872 |
+
"grad_norm": 798713.5625,
|
1873 |
+
"learning_rate": 1.5178096757044127e-05,
|
1874 |
+
"loss": 3.4977,
|
1875 |
+
"step": 26200
|
1876 |
+
},
|
1877 |
+
{
|
1878 |
+
"epoch": 4.1945773524720895,
|
1879 |
+
"grad_norm": 719145.9375,
|
1880 |
+
"learning_rate": 1.5045188729399256e-05,
|
1881 |
+
"loss": 3.4549,
|
1882 |
+
"step": 26300
|
1883 |
+
},
|
1884 |
+
{
|
1885 |
+
"epoch": 4.2105263157894735,
|
1886 |
+
"grad_norm": 692601.9375,
|
1887 |
+
"learning_rate": 1.4912280701754386e-05,
|
1888 |
+
"loss": 3.5228,
|
1889 |
+
"step": 26400
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 4.226475279106858,
|
1893 |
+
"grad_norm": 631408.4375,
|
1894 |
+
"learning_rate": 1.4779372674109517e-05,
|
1895 |
+
"loss": 3.4742,
|
1896 |
+
"step": 26500
|
1897 |
+
},
|
1898 |
+
{
|
1899 |
+
"epoch": 4.242424242424242,
|
1900 |
+
"grad_norm": 733137.0,
|
1901 |
+
"learning_rate": 1.4646464646464647e-05,
|
1902 |
+
"loss": 3.4242,
|
1903 |
+
"step": 26600
|
1904 |
+
},
|
1905 |
+
{
|
1906 |
+
"epoch": 4.258373205741627,
|
1907 |
+
"grad_norm": 741812.9375,
|
1908 |
+
"learning_rate": 1.4513556618819777e-05,
|
1909 |
+
"loss": 3.4598,
|
1910 |
+
"step": 26700
|
1911 |
+
},
|
1912 |
+
{
|
1913 |
+
"epoch": 4.274322169059011,
|
1914 |
+
"grad_norm": 682720.125,
|
1915 |
+
"learning_rate": 1.4380648591174908e-05,
|
1916 |
+
"loss": 3.4827,
|
1917 |
+
"step": 26800
|
1918 |
+
},
|
1919 |
+
{
|
1920 |
+
"epoch": 4.290271132376396,
|
1921 |
+
"grad_norm": 784646.125,
|
1922 |
+
"learning_rate": 1.4247740563530037e-05,
|
1923 |
+
"loss": 3.4712,
|
1924 |
+
"step": 26900
|
1925 |
+
},
|
1926 |
+
{
|
1927 |
+
"epoch": 4.30622009569378,
|
1928 |
+
"grad_norm": 713282.75,
|
1929 |
+
"learning_rate": 1.4114832535885167e-05,
|
1930 |
+
"loss": 3.5542,
|
1931 |
+
"step": 27000
|
1932 |
+
},
|
1933 |
+
{
|
1934 |
+
"epoch": 4.3221690590111645,
|
1935 |
+
"grad_norm": 699440.4375,
|
1936 |
+
"learning_rate": 1.3981924508240298e-05,
|
1937 |
+
"loss": 3.4572,
|
1938 |
+
"step": 27100
|
1939 |
+
},
|
1940 |
+
{
|
1941 |
+
"epoch": 4.3381180223285485,
|
1942 |
+
"grad_norm": 733063.875,
|
1943 |
+
"learning_rate": 1.3849016480595428e-05,
|
1944 |
+
"loss": 3.4106,
|
1945 |
+
"step": 27200
|
1946 |
+
},
|
1947 |
+
{
|
1948 |
+
"epoch": 4.354066985645933,
|
1949 |
+
"grad_norm": 799606.875,
|
1950 |
+
"learning_rate": 1.371610845295056e-05,
|
1951 |
+
"loss": 3.4189,
|
1952 |
+
"step": 27300
|
1953 |
+
},
|
1954 |
+
{
|
1955 |
+
"epoch": 4.370015948963317,
|
1956 |
+
"grad_norm": 722583.25,
|
1957 |
+
"learning_rate": 1.3583200425305689e-05,
|
1958 |
+
"loss": 3.448,
|
1959 |
+
"step": 27400
|
1960 |
+
},
|
1961 |
+
{
|
1962 |
+
"epoch": 4.385964912280702,
|
1963 |
+
"grad_norm": 761491.5625,
|
1964 |
+
"learning_rate": 1.3450292397660819e-05,
|
1965 |
+
"loss": 3.5248,
|
1966 |
+
"step": 27500
|
1967 |
+
},
|
1968 |
+
{
|
1969 |
+
"epoch": 4.401913875598086,
|
1970 |
+
"grad_norm": 633397.6875,
|
1971 |
+
"learning_rate": 1.331738437001595e-05,
|
1972 |
+
"loss": 3.5675,
|
1973 |
+
"step": 27600
|
1974 |
+
},
|
1975 |
+
{
|
1976 |
+
"epoch": 4.417862838915471,
|
1977 |
+
"grad_norm": 743160.4375,
|
1978 |
+
"learning_rate": 1.318447634237108e-05,
|
1979 |
+
"loss": 3.4983,
|
1980 |
+
"step": 27700
|
1981 |
+
},
|
1982 |
+
{
|
1983 |
+
"epoch": 4.433811802232855,
|
1984 |
+
"grad_norm": 689363.3125,
|
1985 |
+
"learning_rate": 1.305156831472621e-05,
|
1986 |
+
"loss": 3.5689,
|
1987 |
+
"step": 27800
|
1988 |
+
},
|
1989 |
+
{
|
1990 |
+
"epoch": 4.44976076555024,
|
1991 |
+
"grad_norm": 634674.0,
|
1992 |
+
"learning_rate": 1.291866028708134e-05,
|
1993 |
+
"loss": 3.4763,
|
1994 |
+
"step": 27900
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 4.4657097288676235,
|
1998 |
+
"grad_norm": 682868.5,
|
1999 |
+
"learning_rate": 1.278575225943647e-05,
|
2000 |
+
"loss": 3.4725,
|
2001 |
+
"step": 28000
|
2002 |
+
},
|
2003 |
+
{
|
2004 |
+
"epoch": 4.481658692185008,
|
2005 |
+
"grad_norm": 839863.0,
|
2006 |
+
"learning_rate": 1.26528442317916e-05,
|
2007 |
+
"loss": 3.5255,
|
2008 |
+
"step": 28100
|
2009 |
+
},
|
2010 |
+
{
|
2011 |
+
"epoch": 4.497607655502392,
|
2012 |
+
"grad_norm": 840497.0625,
|
2013 |
+
"learning_rate": 1.2519936204146731e-05,
|
2014 |
+
"loss": 3.4914,
|
2015 |
+
"step": 28200
|
2016 |
+
},
|
2017 |
+
{
|
2018 |
+
"epoch": 4.513556618819777,
|
2019 |
+
"grad_norm": 798480.8125,
|
2020 |
+
"learning_rate": 1.2387028176501861e-05,
|
2021 |
+
"loss": 3.4787,
|
2022 |
+
"step": 28300
|
2023 |
+
},
|
2024 |
+
{
|
2025 |
+
"epoch": 4.529505582137161,
|
2026 |
+
"grad_norm": 776783.625,
|
2027 |
+
"learning_rate": 1.2254120148856992e-05,
|
2028 |
+
"loss": 3.4817,
|
2029 |
+
"step": 28400
|
2030 |
+
},
|
2031 |
+
{
|
2032 |
+
"epoch": 4.545454545454545,
|
2033 |
+
"grad_norm": 860058.625,
|
2034 |
+
"learning_rate": 1.2121212121212122e-05,
|
2035 |
+
"loss": 3.515,
|
2036 |
+
"step": 28500
|
2037 |
+
},
|
2038 |
+
{
|
2039 |
+
"epoch": 4.56140350877193,
|
2040 |
+
"grad_norm": 797709.0,
|
2041 |
+
"learning_rate": 1.1988304093567252e-05,
|
2042 |
+
"loss": 3.4715,
|
2043 |
+
"step": 28600
|
2044 |
+
},
|
2045 |
+
{
|
2046 |
+
"epoch": 4.577352472089315,
|
2047 |
+
"grad_norm": 686232.0,
|
2048 |
+
"learning_rate": 1.1855396065922381e-05,
|
2049 |
+
"loss": 3.493,
|
2050 |
+
"step": 28700
|
2051 |
+
},
|
2052 |
+
{
|
2053 |
+
"epoch": 4.5933014354066986,
|
2054 |
+
"grad_norm": 713671.875,
|
2055 |
+
"learning_rate": 1.1722488038277513e-05,
|
2056 |
+
"loss": 3.4767,
|
2057 |
+
"step": 28800
|
2058 |
+
},
|
2059 |
+
{
|
2060 |
+
"epoch": 4.6092503987240825,
|
2061 |
+
"grad_norm": 830858.875,
|
2062 |
+
"learning_rate": 1.1589580010632644e-05,
|
2063 |
+
"loss": 3.5506,
|
2064 |
+
"step": 28900
|
2065 |
+
},
|
2066 |
+
{
|
2067 |
+
"epoch": 4.625199362041467,
|
2068 |
+
"grad_norm": 681684.8125,
|
2069 |
+
"learning_rate": 1.1456671982987772e-05,
|
2070 |
+
"loss": 3.468,
|
2071 |
+
"step": 29000
|
2072 |
+
},
|
2073 |
+
{
|
2074 |
+
"epoch": 4.641148325358852,
|
2075 |
+
"grad_norm": 693863.25,
|
2076 |
+
"learning_rate": 1.1323763955342903e-05,
|
2077 |
+
"loss": 3.4847,
|
2078 |
+
"step": 29100
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 4.657097288676236,
|
2082 |
+
"grad_norm": 612233.75,
|
2083 |
+
"learning_rate": 1.1190855927698035e-05,
|
2084 |
+
"loss": 3.3994,
|
2085 |
+
"step": 29200
|
2086 |
+
},
|
2087 |
+
{
|
2088 |
+
"epoch": 4.67304625199362,
|
2089 |
+
"grad_norm": 901251.25,
|
2090 |
+
"learning_rate": 1.1057947900053164e-05,
|
2091 |
+
"loss": 3.5144,
|
2092 |
+
"step": 29300
|
2093 |
+
},
|
2094 |
+
{
|
2095 |
+
"epoch": 4.688995215311005,
|
2096 |
+
"grad_norm": 742618.4375,
|
2097 |
+
"learning_rate": 1.0925039872408294e-05,
|
2098 |
+
"loss": 3.5388,
|
2099 |
+
"step": 29400
|
2100 |
+
},
|
2101 |
+
{
|
2102 |
+
"epoch": 4.70494417862839,
|
2103 |
+
"grad_norm": 797654.0,
|
2104 |
+
"learning_rate": 1.0792131844763423e-05,
|
2105 |
+
"loss": 3.4722,
|
2106 |
+
"step": 29500
|
2107 |
+
},
|
2108 |
+
{
|
2109 |
+
"epoch": 4.720893141945774,
|
2110 |
+
"grad_norm": 763737.9375,
|
2111 |
+
"learning_rate": 1.0659223817118555e-05,
|
2112 |
+
"loss": 3.5006,
|
2113 |
+
"step": 29600
|
2114 |
+
},
|
2115 |
+
{
|
2116 |
+
"epoch": 4.7368421052631575,
|
2117 |
+
"grad_norm": 702353.875,
|
2118 |
+
"learning_rate": 1.0526315789473684e-05,
|
2119 |
+
"loss": 3.4772,
|
2120 |
+
"step": 29700
|
2121 |
+
},
|
2122 |
+
{
|
2123 |
+
"epoch": 4.752791068580542,
|
2124 |
+
"grad_norm": 790482.1875,
|
2125 |
+
"learning_rate": 1.0393407761828814e-05,
|
2126 |
+
"loss": 3.4507,
|
2127 |
+
"step": 29800
|
2128 |
+
},
|
2129 |
+
{
|
2130 |
+
"epoch": 4.768740031897926,
|
2131 |
+
"grad_norm": 706455.9375,
|
2132 |
+
"learning_rate": 1.0260499734183945e-05,
|
2133 |
+
"loss": 3.4769,
|
2134 |
+
"step": 29900
|
2135 |
+
},
|
2136 |
+
{
|
2137 |
+
"epoch": 4.784688995215311,
|
2138 |
+
"grad_norm": 720554.3125,
|
2139 |
+
"learning_rate": 1.0127591706539077e-05,
|
2140 |
+
"loss": 3.4551,
|
2141 |
+
"step": 30000
|
2142 |
+
},
|
2143 |
+
{
|
2144 |
+
"epoch": 4.800637958532695,
|
2145 |
+
"grad_norm": 754827.6875,
|
2146 |
+
"learning_rate": 9.994683678894205e-06,
|
2147 |
+
"loss": 3.4982,
|
2148 |
+
"step": 30100
|
2149 |
+
},
|
2150 |
+
{
|
2151 |
+
"epoch": 4.81658692185008,
|
2152 |
+
"grad_norm": 696089.375,
|
2153 |
+
"learning_rate": 9.861775651249336e-06,
|
2154 |
+
"loss": 3.4637,
|
2155 |
+
"step": 30200
|
2156 |
+
},
|
2157 |
+
{
|
2158 |
+
"epoch": 4.832535885167464,
|
2159 |
+
"grad_norm": 737095.3125,
|
2160 |
+
"learning_rate": 9.728867623604466e-06,
|
2161 |
+
"loss": 3.4621,
|
2162 |
+
"step": 30300
|
2163 |
+
},
|
2164 |
+
{
|
2165 |
+
"epoch": 4.848484848484849,
|
2166 |
+
"grad_norm": 768262.125,
|
2167 |
+
"learning_rate": 9.595959595959595e-06,
|
2168 |
+
"loss": 3.48,
|
2169 |
+
"step": 30400
|
2170 |
+
},
|
2171 |
+
{
|
2172 |
+
"epoch": 4.8644338118022326,
|
2173 |
+
"grad_norm": 767420.5625,
|
2174 |
+
"learning_rate": 9.463051568314727e-06,
|
2175 |
+
"loss": 3.4463,
|
2176 |
+
"step": 30500
|
2177 |
+
},
|
2178 |
+
{
|
2179 |
+
"epoch": 4.880382775119617,
|
2180 |
+
"grad_norm": 706310.0,
|
2181 |
+
"learning_rate": 9.330143540669856e-06,
|
2182 |
+
"loss": 3.4668,
|
2183 |
+
"step": 30600
|
2184 |
+
},
|
2185 |
+
{
|
2186 |
+
"epoch": 4.896331738437001,
|
2187 |
+
"grad_norm": 828940.625,
|
2188 |
+
"learning_rate": 9.197235513024988e-06,
|
2189 |
+
"loss": 3.4377,
|
2190 |
+
"step": 30700
|
2191 |
+
},
|
2192 |
+
{
|
2193 |
+
"epoch": 4.912280701754386,
|
2194 |
+
"grad_norm": 788075.375,
|
2195 |
+
"learning_rate": 9.064327485380117e-06,
|
2196 |
+
"loss": 3.4144,
|
2197 |
+
"step": 30800
|
2198 |
+
},
|
2199 |
+
{
|
2200 |
+
"epoch": 4.92822966507177,
|
2201 |
+
"grad_norm": 805364.0625,
|
2202 |
+
"learning_rate": 8.931419457735247e-06,
|
2203 |
+
"loss": 3.4726,
|
2204 |
+
"step": 30900
|
2205 |
+
},
|
2206 |
+
{
|
2207 |
+
"epoch": 4.944178628389155,
|
2208 |
+
"grad_norm": 604472.75,
|
2209 |
+
"learning_rate": 8.798511430090378e-06,
|
2210 |
+
"loss": 3.4936,
|
2211 |
+
"step": 31000
|
2212 |
+
},
|
2213 |
+
{
|
2214 |
+
"epoch": 4.960127591706539,
|
2215 |
+
"grad_norm": 684835.3125,
|
2216 |
+
"learning_rate": 8.66560340244551e-06,
|
2217 |
+
"loss": 3.4637,
|
2218 |
+
"step": 31100
|
2219 |
+
},
|
2220 |
+
{
|
2221 |
+
"epoch": 4.976076555023924,
|
2222 |
+
"grad_norm": 684998.5,
|
2223 |
+
"learning_rate": 8.532695374800638e-06,
|
2224 |
+
"loss": 3.5321,
|
2225 |
+
"step": 31200
|
2226 |
+
},
|
2227 |
+
{
|
2228 |
+
"epoch": 4.992025518341308,
|
2229 |
+
"grad_norm": 711871.5625,
|
2230 |
+
"learning_rate": 8.399787347155769e-06,
|
2231 |
+
"loss": 3.4809,
|
2232 |
+
"step": 31300
|
2233 |
+
},
|
2234 |
+
{
|
2235 |
+
"epoch": 5.0,
|
2236 |
+
"eval_loss": 3.6019129753112793,
|
2237 |
+
"eval_runtime": 212.6906,
|
2238 |
+
"eval_samples_per_second": 134.576,
|
2239 |
+
"eval_steps_per_second": 4.208,
|
2240 |
+
"step": 31350
|
2241 |
+
}
|
2242 |
+
],
|
2243 |
+
"logging_steps": 100,
|
2244 |
+
"max_steps": 37620,
|
2245 |
+
"num_input_tokens_seen": 0,
|
2246 |
+
"num_train_epochs": 6,
|
2247 |
+
"save_steps": 500,
|
2248 |
+
"stateful_callbacks": {
|
2249 |
+
"TrainerControl": {
|
2250 |
+
"args": {
|
2251 |
+
"should_epoch_stop": false,
|
2252 |
+
"should_evaluate": false,
|
2253 |
+
"should_log": false,
|
2254 |
+
"should_save": true,
|
2255 |
+
"should_training_stop": false
|
2256 |
+
},
|
2257 |
+
"attributes": {}
|
2258 |
+
}
|
2259 |
+
},
|
2260 |
+
"total_flos": 3.881011443758592e+16,
|
2261 |
+
"train_batch_size": 32,
|
2262 |
+
"trial_name": null,
|
2263 |
+
"trial_params": null
|
2264 |
+
}
|
checkpoint-31350/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7e6d8905717aeff0bee4d2b97e0432173098aa3c32a8f4e99a68f0803ee335cd
|
3 |
+
size 5240
|
checkpoint-31350/vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-37620/config.json
ADDED
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "Ransaka/sinhala-bert-medium-v2",
|
3 |
+
"architectures": [
|
4 |
+
"BertForMaskedLM"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_dropout": null,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_dropout_prob": 0.1,
|
10 |
+
"hidden_size": 786,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 1024,
|
13 |
+
"layer_norm_eps": 1e-12,
|
14 |
+
"max_position_embeddings": 256,
|
15 |
+
"model_type": "bert",
|
16 |
+
"num_attention_heads": 6,
|
17 |
+
"num_hidden_layers": 6,
|
18 |
+
"pad_token_id": 0,
|
19 |
+
"position_embedding_type": "absolute",
|
20 |
+
"torch_dtype": "float32",
|
21 |
+
"transformers_version": "4.47.0",
|
22 |
+
"type_vocab_size": 2,
|
23 |
+
"use_cache": true,
|
24 |
+
"vocab_size": 32000
|
25 |
+
}
|
checkpoint-37620/generation_config.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"pad_token_id": 0,
|
4 |
+
"transformers_version": "4.47.0"
|
5 |
+
}
|
checkpoint-37620/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:51874da6e509fec39e26442cbb403925b98087382b3935fe751f465f12a14781
|
3 |
+
size 202182680
|
checkpoint-37620/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2c34063c5c901ee1afd5592c6ffbc9f35c7f259aa87f3a3d73744b818650009e
|
3 |
+
size 404427706
|
checkpoint-37620/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:74143774766502ce979c9960d73e58c82c04649b6fa34eacaf30aafbae0c5aee
|
3 |
+
size 14244
|
checkpoint-37620/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:72f104db74301089ced9d13f69518b106b2a706eabaeae6880252941368f78ab
|
3 |
+
size 1064
|
checkpoint-37620/special_tokens_map.json
ADDED
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": {
|
3 |
+
"content": "[CLS]",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"mask_token": {
|
10 |
+
"content": "[MASK]",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"pad_token": {
|
17 |
+
"content": "[PAD]",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"sep_token": {
|
24 |
+
"content": "[SEP]",
|
25 |
+
"lstrip": false,
|
26 |
+
"normalized": false,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
},
|
30 |
+
"unk_token": {
|
31 |
+
"content": "[UNK]",
|
32 |
+
"lstrip": false,
|
33 |
+
"normalized": false,
|
34 |
+
"rstrip": false,
|
35 |
+
"single_word": false
|
36 |
+
}
|
37 |
+
}
|
checkpoint-37620/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-37620/tokenizer_config.json
ADDED
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "[PAD]",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "[UNK]",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "[CLS]",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "[SEP]",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"4": {
|
36 |
+
"content": "[MASK]",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"clean_up_tokenization_spaces": true,
|
45 |
+
"cls_token": "[CLS]",
|
46 |
+
"do_lower_case": false,
|
47 |
+
"extra_special_tokens": {},
|
48 |
+
"mask_token": "[MASK]",
|
49 |
+
"model_max_length": 512,
|
50 |
+
"pad_token": "[PAD]",
|
51 |
+
"sep_token": "[SEP]",
|
52 |
+
"strip_accents": null,
|
53 |
+
"tokenize_chinese_chars": true,
|
54 |
+
"tokenizer_class": "BertTokenizer",
|
55 |
+
"unk_token": "[UNK]"
|
56 |
+
}
|
checkpoint-37620/trainer_state.json
ADDED
@@ -0,0 +1,2713 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 6.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 37620,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.01594896331738437,
|
13 |
+
"grad_norm": 793786.1875,
|
14 |
+
"learning_rate": 4.9867091972355135e-05,
|
15 |
+
"loss": 4.3231,
|
16 |
+
"step": 100
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.03189792663476874,
|
20 |
+
"grad_norm": 705751.0625,
|
21 |
+
"learning_rate": 4.973418394471026e-05,
|
22 |
+
"loss": 4.1867,
|
23 |
+
"step": 200
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.04784688995215311,
|
27 |
+
"grad_norm": 761077.1875,
|
28 |
+
"learning_rate": 4.9601275917065395e-05,
|
29 |
+
"loss": 4.225,
|
30 |
+
"step": 300
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.06379585326953748,
|
34 |
+
"grad_norm": 660073.9375,
|
35 |
+
"learning_rate": 4.946836788942053e-05,
|
36 |
+
"loss": 4.2529,
|
37 |
+
"step": 400
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.07974481658692185,
|
41 |
+
"grad_norm": 701404.4375,
|
42 |
+
"learning_rate": 4.9335459861775654e-05,
|
43 |
+
"loss": 4.2985,
|
44 |
+
"step": 500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.09569377990430622,
|
48 |
+
"grad_norm": 776949.5625,
|
49 |
+
"learning_rate": 4.920255183413078e-05,
|
50 |
+
"loss": 4.2124,
|
51 |
+
"step": 600
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.11164274322169059,
|
55 |
+
"grad_norm": 703487.375,
|
56 |
+
"learning_rate": 4.906964380648591e-05,
|
57 |
+
"loss": 4.224,
|
58 |
+
"step": 700
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.12759170653907495,
|
62 |
+
"grad_norm": 727441.625,
|
63 |
+
"learning_rate": 4.893673577884104e-05,
|
64 |
+
"loss": 4.3259,
|
65 |
+
"step": 800
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.14354066985645933,
|
69 |
+
"grad_norm": 772143.4375,
|
70 |
+
"learning_rate": 4.880382775119617e-05,
|
71 |
+
"loss": 4.1641,
|
72 |
+
"step": 900
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.1594896331738437,
|
76 |
+
"grad_norm": 834641.6875,
|
77 |
+
"learning_rate": 4.8670919723551306e-05,
|
78 |
+
"loss": 4.2651,
|
79 |
+
"step": 1000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.17543859649122806,
|
83 |
+
"grad_norm": 743223.0,
|
84 |
+
"learning_rate": 4.853801169590643e-05,
|
85 |
+
"loss": 4.1867,
|
86 |
+
"step": 1100
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.19138755980861244,
|
90 |
+
"grad_norm": 643478.125,
|
91 |
+
"learning_rate": 4.8405103668261565e-05,
|
92 |
+
"loss": 4.1816,
|
93 |
+
"step": 1200
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.20733652312599682,
|
97 |
+
"grad_norm": 656405.6875,
|
98 |
+
"learning_rate": 4.82721956406167e-05,
|
99 |
+
"loss": 4.1677,
|
100 |
+
"step": 1300
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.22328548644338117,
|
104 |
+
"grad_norm": 714701.3125,
|
105 |
+
"learning_rate": 4.813928761297183e-05,
|
106 |
+
"loss": 4.1595,
|
107 |
+
"step": 1400
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.23923444976076555,
|
111 |
+
"grad_norm": 739746.875,
|
112 |
+
"learning_rate": 4.800637958532696e-05,
|
113 |
+
"loss": 4.1319,
|
114 |
+
"step": 1500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.2551834130781499,
|
118 |
+
"grad_norm": 736049.3125,
|
119 |
+
"learning_rate": 4.787347155768209e-05,
|
120 |
+
"loss": 4.1543,
|
121 |
+
"step": 1600
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.2711323763955343,
|
125 |
+
"grad_norm": 668736.5,
|
126 |
+
"learning_rate": 4.7740563530037217e-05,
|
127 |
+
"loss": 4.2052,
|
128 |
+
"step": 1700
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.28708133971291866,
|
132 |
+
"grad_norm": 655993.0625,
|
133 |
+
"learning_rate": 4.760765550239234e-05,
|
134 |
+
"loss": 4.1642,
|
135 |
+
"step": 1800
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.30303030303030304,
|
139 |
+
"grad_norm": 732162.3125,
|
140 |
+
"learning_rate": 4.7474747474747476e-05,
|
141 |
+
"loss": 4.1464,
|
142 |
+
"step": 1900
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.3189792663476874,
|
146 |
+
"grad_norm": 687130.875,
|
147 |
+
"learning_rate": 4.734183944710261e-05,
|
148 |
+
"loss": 4.181,
|
149 |
+
"step": 2000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.3349282296650718,
|
153 |
+
"grad_norm": 823373.1875,
|
154 |
+
"learning_rate": 4.7208931419457735e-05,
|
155 |
+
"loss": 4.1591,
|
156 |
+
"step": 2100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.3508771929824561,
|
160 |
+
"grad_norm": 716504.625,
|
161 |
+
"learning_rate": 4.707602339181287e-05,
|
162 |
+
"loss": 4.1391,
|
163 |
+
"step": 2200
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.3668261562998405,
|
167 |
+
"grad_norm": 714200.6875,
|
168 |
+
"learning_rate": 4.6943115364168e-05,
|
169 |
+
"loss": 4.0962,
|
170 |
+
"step": 2300
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.3827751196172249,
|
174 |
+
"grad_norm": 662553.0,
|
175 |
+
"learning_rate": 4.681020733652313e-05,
|
176 |
+
"loss": 4.0964,
|
177 |
+
"step": 2400
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.39872408293460926,
|
181 |
+
"grad_norm": 767714.875,
|
182 |
+
"learning_rate": 4.667729930887826e-05,
|
183 |
+
"loss": 4.1174,
|
184 |
+
"step": 2500
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.41467304625199364,
|
188 |
+
"grad_norm": 763852.75,
|
189 |
+
"learning_rate": 4.6544391281233393e-05,
|
190 |
+
"loss": 4.0442,
|
191 |
+
"step": 2600
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.430622009569378,
|
195 |
+
"grad_norm": 733045.8125,
|
196 |
+
"learning_rate": 4.641148325358852e-05,
|
197 |
+
"loss": 4.112,
|
198 |
+
"step": 2700
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.44657097288676234,
|
202 |
+
"grad_norm": 752065.875,
|
203 |
+
"learning_rate": 4.6278575225943646e-05,
|
204 |
+
"loss": 4.1223,
|
205 |
+
"step": 2800
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.4625199362041467,
|
209 |
+
"grad_norm": 675739.8125,
|
210 |
+
"learning_rate": 4.614566719829878e-05,
|
211 |
+
"loss": 4.1608,
|
212 |
+
"step": 2900
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.4784688995215311,
|
216 |
+
"grad_norm": 666147.375,
|
217 |
+
"learning_rate": 4.6012759170653905e-05,
|
218 |
+
"loss": 4.0803,
|
219 |
+
"step": 3000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.4944178628389155,
|
223 |
+
"grad_norm": 594287.9375,
|
224 |
+
"learning_rate": 4.587985114300904e-05,
|
225 |
+
"loss": 4.045,
|
226 |
+
"step": 3100
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.5103668261562998,
|
230 |
+
"grad_norm": 778230.25,
|
231 |
+
"learning_rate": 4.574694311536417e-05,
|
232 |
+
"loss": 4.1268,
|
233 |
+
"step": 3200
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.5263157894736842,
|
237 |
+
"grad_norm": 683312.0625,
|
238 |
+
"learning_rate": 4.56140350877193e-05,
|
239 |
+
"loss": 4.0424,
|
240 |
+
"step": 3300
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.5422647527910686,
|
244 |
+
"grad_norm": 640858.25,
|
245 |
+
"learning_rate": 4.548112706007443e-05,
|
246 |
+
"loss": 4.0235,
|
247 |
+
"step": 3400
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.5582137161084529,
|
251 |
+
"grad_norm": 680217.875,
|
252 |
+
"learning_rate": 4.5348219032429564e-05,
|
253 |
+
"loss": 4.1423,
|
254 |
+
"step": 3500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.5741626794258373,
|
258 |
+
"grad_norm": 708408.4375,
|
259 |
+
"learning_rate": 4.521531100478469e-05,
|
260 |
+
"loss": 4.0291,
|
261 |
+
"step": 3600
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.5901116427432217,
|
265 |
+
"grad_norm": 663333.0,
|
266 |
+
"learning_rate": 4.508240297713982e-05,
|
267 |
+
"loss": 3.9998,
|
268 |
+
"step": 3700
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.6060606060606061,
|
272 |
+
"grad_norm": 652592.0,
|
273 |
+
"learning_rate": 4.494949494949495e-05,
|
274 |
+
"loss": 4.0424,
|
275 |
+
"step": 3800
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.6220095693779905,
|
279 |
+
"grad_norm": 667416.9375,
|
280 |
+
"learning_rate": 4.481658692185008e-05,
|
281 |
+
"loss": 4.053,
|
282 |
+
"step": 3900
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.6379585326953748,
|
286 |
+
"grad_norm": 674108.4375,
|
287 |
+
"learning_rate": 4.468367889420521e-05,
|
288 |
+
"loss": 4.0748,
|
289 |
+
"step": 4000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.6539074960127592,
|
293 |
+
"grad_norm": 714003.625,
|
294 |
+
"learning_rate": 4.455077086656034e-05,
|
295 |
+
"loss": 3.9807,
|
296 |
+
"step": 4100
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.6698564593301436,
|
300 |
+
"grad_norm": 693166.4375,
|
301 |
+
"learning_rate": 4.4417862838915475e-05,
|
302 |
+
"loss": 3.993,
|
303 |
+
"step": 4200
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.6858054226475279,
|
307 |
+
"grad_norm": 689997.125,
|
308 |
+
"learning_rate": 4.42849548112706e-05,
|
309 |
+
"loss": 4.0593,
|
310 |
+
"step": 4300
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.7017543859649122,
|
314 |
+
"grad_norm": 741118.5625,
|
315 |
+
"learning_rate": 4.4152046783625734e-05,
|
316 |
+
"loss": 4.0409,
|
317 |
+
"step": 4400
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.7177033492822966,
|
321 |
+
"grad_norm": 668626.9375,
|
322 |
+
"learning_rate": 4.401913875598087e-05,
|
323 |
+
"loss": 4.06,
|
324 |
+
"step": 4500
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.733652312599681,
|
328 |
+
"grad_norm": 771064.125,
|
329 |
+
"learning_rate": 4.388623072833599e-05,
|
330 |
+
"loss": 4.0122,
|
331 |
+
"step": 4600
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.7496012759170654,
|
335 |
+
"grad_norm": 756131.375,
|
336 |
+
"learning_rate": 4.3753322700691126e-05,
|
337 |
+
"loss": 4.043,
|
338 |
+
"step": 4700
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.7655502392344498,
|
342 |
+
"grad_norm": 737817.5625,
|
343 |
+
"learning_rate": 4.362041467304626e-05,
|
344 |
+
"loss": 4.0213,
|
345 |
+
"step": 4800
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.7814992025518341,
|
349 |
+
"grad_norm": 758023.5625,
|
350 |
+
"learning_rate": 4.3487506645401385e-05,
|
351 |
+
"loss": 3.9778,
|
352 |
+
"step": 4900
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.7974481658692185,
|
356 |
+
"grad_norm": 672895.1875,
|
357 |
+
"learning_rate": 4.335459861775651e-05,
|
358 |
+
"loss": 4.0434,
|
359 |
+
"step": 5000
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.8133971291866029,
|
363 |
+
"grad_norm": 710937.375,
|
364 |
+
"learning_rate": 4.3221690590111645e-05,
|
365 |
+
"loss": 4.0117,
|
366 |
+
"step": 5100
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.8293460925039873,
|
370 |
+
"grad_norm": 773563.875,
|
371 |
+
"learning_rate": 4.308878256246677e-05,
|
372 |
+
"loss": 4.0516,
|
373 |
+
"step": 5200
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.8452950558213717,
|
377 |
+
"grad_norm": 750733.4375,
|
378 |
+
"learning_rate": 4.2955874534821904e-05,
|
379 |
+
"loss": 3.9798,
|
380 |
+
"step": 5300
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.861244019138756,
|
384 |
+
"grad_norm": 706317.3125,
|
385 |
+
"learning_rate": 4.282296650717704e-05,
|
386 |
+
"loss": 3.9845,
|
387 |
+
"step": 5400
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.8771929824561403,
|
391 |
+
"grad_norm": 710855.75,
|
392 |
+
"learning_rate": 4.269005847953216e-05,
|
393 |
+
"loss": 4.0295,
|
394 |
+
"step": 5500
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.8931419457735247,
|
398 |
+
"grad_norm": 712988.9375,
|
399 |
+
"learning_rate": 4.2557150451887296e-05,
|
400 |
+
"loss": 3.9732,
|
401 |
+
"step": 5600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.9090909090909091,
|
405 |
+
"grad_norm": 628492.75,
|
406 |
+
"learning_rate": 4.242424242424243e-05,
|
407 |
+
"loss": 4.0012,
|
408 |
+
"step": 5700
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.9250398724082934,
|
412 |
+
"grad_norm": 821738.375,
|
413 |
+
"learning_rate": 4.2291334396597556e-05,
|
414 |
+
"loss": 3.9867,
|
415 |
+
"step": 5800
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.9409888357256778,
|
419 |
+
"grad_norm": 720818.125,
|
420 |
+
"learning_rate": 4.215842636895269e-05,
|
421 |
+
"loss": 3.962,
|
422 |
+
"step": 5900
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.9569377990430622,
|
426 |
+
"grad_norm": 698428.0625,
|
427 |
+
"learning_rate": 4.2025518341307815e-05,
|
428 |
+
"loss": 4.0095,
|
429 |
+
"step": 6000
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.9728867623604466,
|
433 |
+
"grad_norm": 769185.75,
|
434 |
+
"learning_rate": 4.189261031366295e-05,
|
435 |
+
"loss": 3.9469,
|
436 |
+
"step": 6100
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.988835725677831,
|
440 |
+
"grad_norm": 706422.25,
|
441 |
+
"learning_rate": 4.1759702286018074e-05,
|
442 |
+
"loss": 3.9466,
|
443 |
+
"step": 6200
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 1.0,
|
447 |
+
"eval_loss": 3.8981289863586426,
|
448 |
+
"eval_runtime": 213.136,
|
449 |
+
"eval_samples_per_second": 134.295,
|
450 |
+
"eval_steps_per_second": 4.199,
|
451 |
+
"step": 6270
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 1.0047846889952152,
|
455 |
+
"grad_norm": 664245.5,
|
456 |
+
"learning_rate": 4.162679425837321e-05,
|
457 |
+
"loss": 3.9397,
|
458 |
+
"step": 6300
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 1.0207336523125996,
|
462 |
+
"grad_norm": 790251.5625,
|
463 |
+
"learning_rate": 4.149388623072834e-05,
|
464 |
+
"loss": 3.7918,
|
465 |
+
"step": 6400
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 1.036682615629984,
|
469 |
+
"grad_norm": 813129.1875,
|
470 |
+
"learning_rate": 4.1360978203083467e-05,
|
471 |
+
"loss": 3.8563,
|
472 |
+
"step": 6500
|
473 |
+
},
|
474 |
+
{
|
475 |
+
"epoch": 1.0526315789473684,
|
476 |
+
"grad_norm": 695871.5,
|
477 |
+
"learning_rate": 4.12280701754386e-05,
|
478 |
+
"loss": 3.959,
|
479 |
+
"step": 6600
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 1.0685805422647527,
|
483 |
+
"grad_norm": 640218.3125,
|
484 |
+
"learning_rate": 4.109516214779373e-05,
|
485 |
+
"loss": 3.9882,
|
486 |
+
"step": 6700
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 1.0845295055821371,
|
490 |
+
"grad_norm": 709886.125,
|
491 |
+
"learning_rate": 4.096225412014886e-05,
|
492 |
+
"loss": 3.9328,
|
493 |
+
"step": 6800
|
494 |
+
},
|
495 |
+
{
|
496 |
+
"epoch": 1.1004784688995215,
|
497 |
+
"grad_norm": 663732.1875,
|
498 |
+
"learning_rate": 4.082934609250399e-05,
|
499 |
+
"loss": 3.947,
|
500 |
+
"step": 6900
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 1.1164274322169059,
|
504 |
+
"grad_norm": 649781.9375,
|
505 |
+
"learning_rate": 4.069643806485912e-05,
|
506 |
+
"loss": 3.8735,
|
507 |
+
"step": 7000
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 1.1323763955342903,
|
511 |
+
"grad_norm": 788934.4375,
|
512 |
+
"learning_rate": 4.056353003721425e-05,
|
513 |
+
"loss": 3.8548,
|
514 |
+
"step": 7100
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 1.1483253588516746,
|
518 |
+
"grad_norm": 688279.3125,
|
519 |
+
"learning_rate": 4.043062200956938e-05,
|
520 |
+
"loss": 3.8844,
|
521 |
+
"step": 7200
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 1.164274322169059,
|
525 |
+
"grad_norm": 698958.0,
|
526 |
+
"learning_rate": 4.029771398192451e-05,
|
527 |
+
"loss": 3.878,
|
528 |
+
"step": 7300
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 1.1802232854864434,
|
532 |
+
"grad_norm": 777544.8125,
|
533 |
+
"learning_rate": 4.016480595427964e-05,
|
534 |
+
"loss": 3.8299,
|
535 |
+
"step": 7400
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 1.1961722488038278,
|
539 |
+
"grad_norm": 768888.3125,
|
540 |
+
"learning_rate": 4.003189792663477e-05,
|
541 |
+
"loss": 3.859,
|
542 |
+
"step": 7500
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 1.2121212121212122,
|
546 |
+
"grad_norm": 703599.8125,
|
547 |
+
"learning_rate": 3.98989898989899e-05,
|
548 |
+
"loss": 3.9008,
|
549 |
+
"step": 7600
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 1.2280701754385965,
|
553 |
+
"grad_norm": 684616.75,
|
554 |
+
"learning_rate": 3.976608187134503e-05,
|
555 |
+
"loss": 3.8917,
|
556 |
+
"step": 7700
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 1.244019138755981,
|
560 |
+
"grad_norm": 754650.9375,
|
561 |
+
"learning_rate": 3.963317384370016e-05,
|
562 |
+
"loss": 3.7987,
|
563 |
+
"step": 7800
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 1.2599681020733653,
|
567 |
+
"grad_norm": 833289.8125,
|
568 |
+
"learning_rate": 3.9500265816055295e-05,
|
569 |
+
"loss": 3.8097,
|
570 |
+
"step": 7900
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 1.2759170653907497,
|
574 |
+
"grad_norm": 712939.125,
|
575 |
+
"learning_rate": 3.936735778841042e-05,
|
576 |
+
"loss": 3.91,
|
577 |
+
"step": 8000
|
578 |
+
},
|
579 |
+
{
|
580 |
+
"epoch": 1.291866028708134,
|
581 |
+
"grad_norm": 691034.4375,
|
582 |
+
"learning_rate": 3.9234449760765554e-05,
|
583 |
+
"loss": 3.8607,
|
584 |
+
"step": 8100
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 1.3078149920255182,
|
588 |
+
"grad_norm": 777656.1875,
|
589 |
+
"learning_rate": 3.910154173312068e-05,
|
590 |
+
"loss": 3.8294,
|
591 |
+
"step": 8200
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 1.3237639553429026,
|
595 |
+
"grad_norm": 779268.6875,
|
596 |
+
"learning_rate": 3.896863370547581e-05,
|
597 |
+
"loss": 3.839,
|
598 |
+
"step": 8300
|
599 |
+
},
|
600 |
+
{
|
601 |
+
"epoch": 1.339712918660287,
|
602 |
+
"grad_norm": 737519.0,
|
603 |
+
"learning_rate": 3.883572567783094e-05,
|
604 |
+
"loss": 3.8773,
|
605 |
+
"step": 8400
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 1.3556618819776713,
|
609 |
+
"grad_norm": 722279.3125,
|
610 |
+
"learning_rate": 3.870281765018607e-05,
|
611 |
+
"loss": 3.8351,
|
612 |
+
"step": 8500
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 1.3716108452950557,
|
616 |
+
"grad_norm": 763525.9375,
|
617 |
+
"learning_rate": 3.8569909622541206e-05,
|
618 |
+
"loss": 3.8517,
|
619 |
+
"step": 8600
|
620 |
+
},
|
621 |
+
{
|
622 |
+
"epoch": 1.38755980861244,
|
623 |
+
"grad_norm": 715289.3125,
|
624 |
+
"learning_rate": 3.843700159489633e-05,
|
625 |
+
"loss": 3.8272,
|
626 |
+
"step": 8700
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 1.4035087719298245,
|
630 |
+
"grad_norm": 656186.625,
|
631 |
+
"learning_rate": 3.8304093567251465e-05,
|
632 |
+
"loss": 3.8894,
|
633 |
+
"step": 8800
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 1.4194577352472089,
|
637 |
+
"grad_norm": 756528.875,
|
638 |
+
"learning_rate": 3.81711855396066e-05,
|
639 |
+
"loss": 3.8387,
|
640 |
+
"step": 8900
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 1.4354066985645932,
|
644 |
+
"grad_norm": 716015.0,
|
645 |
+
"learning_rate": 3.8038277511961725e-05,
|
646 |
+
"loss": 3.7842,
|
647 |
+
"step": 9000
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 1.4513556618819776,
|
651 |
+
"grad_norm": 721006.3125,
|
652 |
+
"learning_rate": 3.790536948431686e-05,
|
653 |
+
"loss": 3.7769,
|
654 |
+
"step": 9100
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 1.467304625199362,
|
658 |
+
"grad_norm": 800538.1875,
|
659 |
+
"learning_rate": 3.7772461456671984e-05,
|
660 |
+
"loss": 3.8393,
|
661 |
+
"step": 9200
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 1.4832535885167464,
|
665 |
+
"grad_norm": 699156.0,
|
666 |
+
"learning_rate": 3.763955342902711e-05,
|
667 |
+
"loss": 3.7527,
|
668 |
+
"step": 9300
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 1.4992025518341308,
|
672 |
+
"grad_norm": 699306.375,
|
673 |
+
"learning_rate": 3.750664540138224e-05,
|
674 |
+
"loss": 3.8306,
|
675 |
+
"step": 9400
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 1.5151515151515151,
|
679 |
+
"grad_norm": 689892.5,
|
680 |
+
"learning_rate": 3.7373737373737376e-05,
|
681 |
+
"loss": 3.8582,
|
682 |
+
"step": 9500
|
683 |
+
},
|
684 |
+
{
|
685 |
+
"epoch": 1.5311004784688995,
|
686 |
+
"grad_norm": 712134.3125,
|
687 |
+
"learning_rate": 3.72408293460925e-05,
|
688 |
+
"loss": 3.9048,
|
689 |
+
"step": 9600
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 1.547049441786284,
|
693 |
+
"grad_norm": 611158.625,
|
694 |
+
"learning_rate": 3.7107921318447635e-05,
|
695 |
+
"loss": 3.8315,
|
696 |
+
"step": 9700
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 1.5629984051036683,
|
700 |
+
"grad_norm": 814951.875,
|
701 |
+
"learning_rate": 3.697501329080277e-05,
|
702 |
+
"loss": 3.8433,
|
703 |
+
"step": 9800
|
704 |
+
},
|
705 |
+
{
|
706 |
+
"epoch": 1.5789473684210527,
|
707 |
+
"grad_norm": 638232.25,
|
708 |
+
"learning_rate": 3.6842105263157895e-05,
|
709 |
+
"loss": 3.8697,
|
710 |
+
"step": 9900
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 1.594896331738437,
|
714 |
+
"grad_norm": 717277.8125,
|
715 |
+
"learning_rate": 3.670919723551303e-05,
|
716 |
+
"loss": 3.8447,
|
717 |
+
"step": 10000
|
718 |
+
},
|
719 |
+
{
|
720 |
+
"epoch": 1.6108452950558214,
|
721 |
+
"grad_norm": 623000.1875,
|
722 |
+
"learning_rate": 3.657628920786816e-05,
|
723 |
+
"loss": 3.9104,
|
724 |
+
"step": 10100
|
725 |
+
},
|
726 |
+
{
|
727 |
+
"epoch": 1.6267942583732058,
|
728 |
+
"grad_norm": 737050.875,
|
729 |
+
"learning_rate": 3.644338118022329e-05,
|
730 |
+
"loss": 3.8208,
|
731 |
+
"step": 10200
|
732 |
+
},
|
733 |
+
{
|
734 |
+
"epoch": 1.6427432216905902,
|
735 |
+
"grad_norm": 766050.0,
|
736 |
+
"learning_rate": 3.631047315257842e-05,
|
737 |
+
"loss": 3.8454,
|
738 |
+
"step": 10300
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 1.6586921850079746,
|
742 |
+
"grad_norm": 681873.25,
|
743 |
+
"learning_rate": 3.6177565124933546e-05,
|
744 |
+
"loss": 3.8284,
|
745 |
+
"step": 10400
|
746 |
+
},
|
747 |
+
{
|
748 |
+
"epoch": 1.674641148325359,
|
749 |
+
"grad_norm": 636321.0625,
|
750 |
+
"learning_rate": 3.604465709728867e-05,
|
751 |
+
"loss": 3.8191,
|
752 |
+
"step": 10500
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 1.6905901116427433,
|
756 |
+
"grad_norm": 751694.875,
|
757 |
+
"learning_rate": 3.5911749069643806e-05,
|
758 |
+
"loss": 3.8152,
|
759 |
+
"step": 10600
|
760 |
+
},
|
761 |
+
{
|
762 |
+
"epoch": 1.7065390749601277,
|
763 |
+
"grad_norm": 779390.25,
|
764 |
+
"learning_rate": 3.577884104199894e-05,
|
765 |
+
"loss": 3.8235,
|
766 |
+
"step": 10700
|
767 |
+
},
|
768 |
+
{
|
769 |
+
"epoch": 1.722488038277512,
|
770 |
+
"grad_norm": 780193.4375,
|
771 |
+
"learning_rate": 3.5645933014354065e-05,
|
772 |
+
"loss": 3.8109,
|
773 |
+
"step": 10800
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 1.7384370015948964,
|
777 |
+
"grad_norm": 678532.75,
|
778 |
+
"learning_rate": 3.55130249867092e-05,
|
779 |
+
"loss": 3.8017,
|
780 |
+
"step": 10900
|
781 |
+
},
|
782 |
+
{
|
783 |
+
"epoch": 1.7543859649122808,
|
784 |
+
"grad_norm": 733045.375,
|
785 |
+
"learning_rate": 3.538011695906433e-05,
|
786 |
+
"loss": 3.8518,
|
787 |
+
"step": 11000
|
788 |
+
},
|
789 |
+
{
|
790 |
+
"epoch": 1.7703349282296652,
|
791 |
+
"grad_norm": 646417.375,
|
792 |
+
"learning_rate": 3.5247208931419464e-05,
|
793 |
+
"loss": 3.7623,
|
794 |
+
"step": 11100
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 1.7862838915470496,
|
798 |
+
"grad_norm": 641868.25,
|
799 |
+
"learning_rate": 3.511430090377459e-05,
|
800 |
+
"loss": 3.7536,
|
801 |
+
"step": 11200
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 1.802232854864434,
|
805 |
+
"grad_norm": 743804.375,
|
806 |
+
"learning_rate": 3.498139287612972e-05,
|
807 |
+
"loss": 3.8056,
|
808 |
+
"step": 11300
|
809 |
+
},
|
810 |
+
{
|
811 |
+
"epoch": 1.8181818181818183,
|
812 |
+
"grad_norm": 725540.5625,
|
813 |
+
"learning_rate": 3.484848484848485e-05,
|
814 |
+
"loss": 3.8078,
|
815 |
+
"step": 11400
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 1.8341307814992025,
|
819 |
+
"grad_norm": 735264.125,
|
820 |
+
"learning_rate": 3.4715576820839976e-05,
|
821 |
+
"loss": 3.8247,
|
822 |
+
"step": 11500
|
823 |
+
},
|
824 |
+
{
|
825 |
+
"epoch": 1.8500797448165869,
|
826 |
+
"grad_norm": 708785.5625,
|
827 |
+
"learning_rate": 3.458266879319511e-05,
|
828 |
+
"loss": 3.7898,
|
829 |
+
"step": 11600
|
830 |
+
},
|
831 |
+
{
|
832 |
+
"epoch": 1.8660287081339713,
|
833 |
+
"grad_norm": 864416.3125,
|
834 |
+
"learning_rate": 3.444976076555024e-05,
|
835 |
+
"loss": 3.8642,
|
836 |
+
"step": 11700
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 1.8819776714513556,
|
840 |
+
"grad_norm": 693382.5625,
|
841 |
+
"learning_rate": 3.431685273790537e-05,
|
842 |
+
"loss": 3.7433,
|
843 |
+
"step": 11800
|
844 |
+
},
|
845 |
+
{
|
846 |
+
"epoch": 1.89792663476874,
|
847 |
+
"grad_norm": 841755.25,
|
848 |
+
"learning_rate": 3.41839447102605e-05,
|
849 |
+
"loss": 3.8562,
|
850 |
+
"step": 11900
|
851 |
+
},
|
852 |
+
{
|
853 |
+
"epoch": 1.9138755980861244,
|
854 |
+
"grad_norm": 662765.125,
|
855 |
+
"learning_rate": 3.4051036682615634e-05,
|
856 |
+
"loss": 3.8703,
|
857 |
+
"step": 12000
|
858 |
+
},
|
859 |
+
{
|
860 |
+
"epoch": 1.9298245614035088,
|
861 |
+
"grad_norm": 707709.75,
|
862 |
+
"learning_rate": 3.391812865497076e-05,
|
863 |
+
"loss": 3.8402,
|
864 |
+
"step": 12100
|
865 |
+
},
|
866 |
+
{
|
867 |
+
"epoch": 1.9457735247208932,
|
868 |
+
"grad_norm": 616722.875,
|
869 |
+
"learning_rate": 3.3785220627325893e-05,
|
870 |
+
"loss": 3.8432,
|
871 |
+
"step": 12200
|
872 |
+
},
|
873 |
+
{
|
874 |
+
"epoch": 1.9617224880382775,
|
875 |
+
"grad_norm": 795238.625,
|
876 |
+
"learning_rate": 3.3652312599681026e-05,
|
877 |
+
"loss": 3.7319,
|
878 |
+
"step": 12300
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 1.977671451355662,
|
882 |
+
"grad_norm": 709130.0,
|
883 |
+
"learning_rate": 3.351940457203615e-05,
|
884 |
+
"loss": 3.7628,
|
885 |
+
"step": 12400
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 1.9936204146730463,
|
889 |
+
"grad_norm": 789321.375,
|
890 |
+
"learning_rate": 3.3386496544391286e-05,
|
891 |
+
"loss": 3.7821,
|
892 |
+
"step": 12500
|
893 |
+
},
|
894 |
+
{
|
895 |
+
"epoch": 2.0,
|
896 |
+
"eval_loss": 3.800079584121704,
|
897 |
+
"eval_runtime": 212.9573,
|
898 |
+
"eval_samples_per_second": 134.407,
|
899 |
+
"eval_steps_per_second": 4.203,
|
900 |
+
"step": 12540
|
901 |
+
},
|
902 |
+
{
|
903 |
+
"epoch": 2.0095693779904304,
|
904 |
+
"grad_norm": 862786.0,
|
905 |
+
"learning_rate": 3.325358851674641e-05,
|
906 |
+
"loss": 3.7512,
|
907 |
+
"step": 12600
|
908 |
+
},
|
909 |
+
{
|
910 |
+
"epoch": 2.025518341307815,
|
911 |
+
"grad_norm": 772642.5625,
|
912 |
+
"learning_rate": 3.312068048910154e-05,
|
913 |
+
"loss": 3.6611,
|
914 |
+
"step": 12700
|
915 |
+
},
|
916 |
+
{
|
917 |
+
"epoch": 2.041467304625199,
|
918 |
+
"grad_norm": 706907.1875,
|
919 |
+
"learning_rate": 3.298777246145667e-05,
|
920 |
+
"loss": 3.7484,
|
921 |
+
"step": 12800
|
922 |
+
},
|
923 |
+
{
|
924 |
+
"epoch": 2.0574162679425836,
|
925 |
+
"grad_norm": 940896.375,
|
926 |
+
"learning_rate": 3.2854864433811804e-05,
|
927 |
+
"loss": 3.7545,
|
928 |
+
"step": 12900
|
929 |
+
},
|
930 |
+
{
|
931 |
+
"epoch": 2.073365231259968,
|
932 |
+
"grad_norm": 670013.875,
|
933 |
+
"learning_rate": 3.272195640616693e-05,
|
934 |
+
"loss": 3.6504,
|
935 |
+
"step": 13000
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 2.0893141945773523,
|
939 |
+
"grad_norm": 702168.0,
|
940 |
+
"learning_rate": 3.2589048378522064e-05,
|
941 |
+
"loss": 3.6133,
|
942 |
+
"step": 13100
|
943 |
+
},
|
944 |
+
{
|
945 |
+
"epoch": 2.1052631578947367,
|
946 |
+
"grad_norm": 741310.75,
|
947 |
+
"learning_rate": 3.24561403508772e-05,
|
948 |
+
"loss": 3.7202,
|
949 |
+
"step": 13200
|
950 |
+
},
|
951 |
+
{
|
952 |
+
"epoch": 2.121212121212121,
|
953 |
+
"grad_norm": 630812.625,
|
954 |
+
"learning_rate": 3.232323232323233e-05,
|
955 |
+
"loss": 3.7131,
|
956 |
+
"step": 13300
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 2.1371610845295055,
|
960 |
+
"grad_norm": 736768.125,
|
961 |
+
"learning_rate": 3.2190324295587456e-05,
|
962 |
+
"loss": 3.6982,
|
963 |
+
"step": 13400
|
964 |
+
},
|
965 |
+
{
|
966 |
+
"epoch": 2.15311004784689,
|
967 |
+
"grad_norm": 743122.8125,
|
968 |
+
"learning_rate": 3.205741626794259e-05,
|
969 |
+
"loss": 3.7196,
|
970 |
+
"step": 13500
|
971 |
+
},
|
972 |
+
{
|
973 |
+
"epoch": 2.1690590111642742,
|
974 |
+
"grad_norm": 685713.25,
|
975 |
+
"learning_rate": 3.1924508240297715e-05,
|
976 |
+
"loss": 3.7312,
|
977 |
+
"step": 13600
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 2.1850079744816586,
|
981 |
+
"grad_norm": 829695.75,
|
982 |
+
"learning_rate": 3.179160021265284e-05,
|
983 |
+
"loss": 3.6914,
|
984 |
+
"step": 13700
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 2.200956937799043,
|
988 |
+
"grad_norm": 640866.6875,
|
989 |
+
"learning_rate": 3.1658692185007975e-05,
|
990 |
+
"loss": 3.8376,
|
991 |
+
"step": 13800
|
992 |
+
},
|
993 |
+
{
|
994 |
+
"epoch": 2.2169059011164274,
|
995 |
+
"grad_norm": 705478.25,
|
996 |
+
"learning_rate": 3.152578415736311e-05,
|
997 |
+
"loss": 3.7294,
|
998 |
+
"step": 13900
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 2.2328548644338118,
|
1002 |
+
"grad_norm": 664668.8125,
|
1003 |
+
"learning_rate": 3.1392876129718234e-05,
|
1004 |
+
"loss": 3.7046,
|
1005 |
+
"step": 14000
|
1006 |
+
},
|
1007 |
+
{
|
1008 |
+
"epoch": 2.248803827751196,
|
1009 |
+
"grad_norm": 653658.25,
|
1010 |
+
"learning_rate": 3.125996810207337e-05,
|
1011 |
+
"loss": 3.7142,
|
1012 |
+
"step": 14100
|
1013 |
+
},
|
1014 |
+
{
|
1015 |
+
"epoch": 2.2647527910685805,
|
1016 |
+
"grad_norm": 737460.0625,
|
1017 |
+
"learning_rate": 3.11270600744285e-05,
|
1018 |
+
"loss": 3.6449,
|
1019 |
+
"step": 14200
|
1020 |
+
},
|
1021 |
+
{
|
1022 |
+
"epoch": 2.280701754385965,
|
1023 |
+
"grad_norm": 657366.8125,
|
1024 |
+
"learning_rate": 3.0994152046783626e-05,
|
1025 |
+
"loss": 3.7051,
|
1026 |
+
"step": 14300
|
1027 |
+
},
|
1028 |
+
{
|
1029 |
+
"epoch": 2.2966507177033493,
|
1030 |
+
"grad_norm": 734922.0625,
|
1031 |
+
"learning_rate": 3.086124401913876e-05,
|
1032 |
+
"loss": 3.6793,
|
1033 |
+
"step": 14400
|
1034 |
+
},
|
1035 |
+
{
|
1036 |
+
"epoch": 2.3125996810207337,
|
1037 |
+
"grad_norm": 636888.125,
|
1038 |
+
"learning_rate": 3.072833599149389e-05,
|
1039 |
+
"loss": 3.764,
|
1040 |
+
"step": 14500
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 2.328548644338118,
|
1044 |
+
"grad_norm": 776567.125,
|
1045 |
+
"learning_rate": 3.059542796384902e-05,
|
1046 |
+
"loss": 3.6952,
|
1047 |
+
"step": 14600
|
1048 |
+
},
|
1049 |
+
{
|
1050 |
+
"epoch": 2.3444976076555024,
|
1051 |
+
"grad_norm": 678031.5,
|
1052 |
+
"learning_rate": 3.0462519936204148e-05,
|
1053 |
+
"loss": 3.7086,
|
1054 |
+
"step": 14700
|
1055 |
+
},
|
1056 |
+
{
|
1057 |
+
"epoch": 2.360446570972887,
|
1058 |
+
"grad_norm": 702883.9375,
|
1059 |
+
"learning_rate": 3.032961190855928e-05,
|
1060 |
+
"loss": 3.6974,
|
1061 |
+
"step": 14800
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 2.376395534290271,
|
1065 |
+
"grad_norm": 640335.375,
|
1066 |
+
"learning_rate": 3.0196703880914407e-05,
|
1067 |
+
"loss": 3.6555,
|
1068 |
+
"step": 14900
|
1069 |
+
},
|
1070 |
+
{
|
1071 |
+
"epoch": 2.3923444976076556,
|
1072 |
+
"grad_norm": 686070.0625,
|
1073 |
+
"learning_rate": 3.0063795853269537e-05,
|
1074 |
+
"loss": 3.7167,
|
1075 |
+
"step": 15000
|
1076 |
+
},
|
1077 |
+
{
|
1078 |
+
"epoch": 2.40829346092504,
|
1079 |
+
"grad_norm": 676044.25,
|
1080 |
+
"learning_rate": 2.993088782562467e-05,
|
1081 |
+
"loss": 3.6554,
|
1082 |
+
"step": 15100
|
1083 |
+
},
|
1084 |
+
{
|
1085 |
+
"epoch": 2.4242424242424243,
|
1086 |
+
"grad_norm": 767818.6875,
|
1087 |
+
"learning_rate": 2.9797979797979796e-05,
|
1088 |
+
"loss": 3.693,
|
1089 |
+
"step": 15200
|
1090 |
+
},
|
1091 |
+
{
|
1092 |
+
"epoch": 2.4401913875598087,
|
1093 |
+
"grad_norm": 739911.0625,
|
1094 |
+
"learning_rate": 2.966507177033493e-05,
|
1095 |
+
"loss": 3.6431,
|
1096 |
+
"step": 15300
|
1097 |
+
},
|
1098 |
+
{
|
1099 |
+
"epoch": 2.456140350877193,
|
1100 |
+
"grad_norm": 771256.3125,
|
1101 |
+
"learning_rate": 2.9532163742690062e-05,
|
1102 |
+
"loss": 3.6216,
|
1103 |
+
"step": 15400
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 2.4720893141945774,
|
1107 |
+
"grad_norm": 773173.375,
|
1108 |
+
"learning_rate": 2.939925571504519e-05,
|
1109 |
+
"loss": 3.7742,
|
1110 |
+
"step": 15500
|
1111 |
+
},
|
1112 |
+
{
|
1113 |
+
"epoch": 2.488038277511962,
|
1114 |
+
"grad_norm": 762888.3125,
|
1115 |
+
"learning_rate": 2.9266347687400318e-05,
|
1116 |
+
"loss": 3.7236,
|
1117 |
+
"step": 15600
|
1118 |
+
},
|
1119 |
+
{
|
1120 |
+
"epoch": 2.503987240829346,
|
1121 |
+
"grad_norm": 879857.0625,
|
1122 |
+
"learning_rate": 2.913343965975545e-05,
|
1123 |
+
"loss": 3.7253,
|
1124 |
+
"step": 15700
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 2.5199362041467306,
|
1128 |
+
"grad_norm": 733285.0625,
|
1129 |
+
"learning_rate": 2.9000531632110584e-05,
|
1130 |
+
"loss": 3.7166,
|
1131 |
+
"step": 15800
|
1132 |
+
},
|
1133 |
+
{
|
1134 |
+
"epoch": 2.535885167464115,
|
1135 |
+
"grad_norm": 833014.3125,
|
1136 |
+
"learning_rate": 2.886762360446571e-05,
|
1137 |
+
"loss": 3.6705,
|
1138 |
+
"step": 15900
|
1139 |
+
},
|
1140 |
+
{
|
1141 |
+
"epoch": 2.5518341307814993,
|
1142 |
+
"grad_norm": 736943.4375,
|
1143 |
+
"learning_rate": 2.8734715576820844e-05,
|
1144 |
+
"loss": 3.7253,
|
1145 |
+
"step": 16000
|
1146 |
+
},
|
1147 |
+
{
|
1148 |
+
"epoch": 2.5677830940988837,
|
1149 |
+
"grad_norm": 708644.4375,
|
1150 |
+
"learning_rate": 2.8601807549175973e-05,
|
1151 |
+
"loss": 3.6672,
|
1152 |
+
"step": 16100
|
1153 |
+
},
|
1154 |
+
{
|
1155 |
+
"epoch": 2.583732057416268,
|
1156 |
+
"grad_norm": 795735.8125,
|
1157 |
+
"learning_rate": 2.84688995215311e-05,
|
1158 |
+
"loss": 3.728,
|
1159 |
+
"step": 16200
|
1160 |
+
},
|
1161 |
+
{
|
1162 |
+
"epoch": 2.5996810207336525,
|
1163 |
+
"grad_norm": 864533.875,
|
1164 |
+
"learning_rate": 2.8335991493886233e-05,
|
1165 |
+
"loss": 3.6478,
|
1166 |
+
"step": 16300
|
1167 |
+
},
|
1168 |
+
{
|
1169 |
+
"epoch": 2.6156299840510364,
|
1170 |
+
"grad_norm": 835745.0,
|
1171 |
+
"learning_rate": 2.8203083466241366e-05,
|
1172 |
+
"loss": 3.6204,
|
1173 |
+
"step": 16400
|
1174 |
+
},
|
1175 |
+
{
|
1176 |
+
"epoch": 2.6315789473684212,
|
1177 |
+
"grad_norm": 726415.3125,
|
1178 |
+
"learning_rate": 2.8070175438596492e-05,
|
1179 |
+
"loss": 3.6899,
|
1180 |
+
"step": 16500
|
1181 |
+
},
|
1182 |
+
{
|
1183 |
+
"epoch": 2.647527910685805,
|
1184 |
+
"grad_norm": 619188.1875,
|
1185 |
+
"learning_rate": 2.793726741095162e-05,
|
1186 |
+
"loss": 3.6442,
|
1187 |
+
"step": 16600
|
1188 |
+
},
|
1189 |
+
{
|
1190 |
+
"epoch": 2.66347687400319,
|
1191 |
+
"grad_norm": 704278.1875,
|
1192 |
+
"learning_rate": 2.7804359383306755e-05,
|
1193 |
+
"loss": 3.7113,
|
1194 |
+
"step": 16700
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 2.679425837320574,
|
1198 |
+
"grad_norm": 787753.0,
|
1199 |
+
"learning_rate": 2.767145135566188e-05,
|
1200 |
+
"loss": 3.7368,
|
1201 |
+
"step": 16800
|
1202 |
+
},
|
1203 |
+
{
|
1204 |
+
"epoch": 2.6953748006379588,
|
1205 |
+
"grad_norm": 738743.5625,
|
1206 |
+
"learning_rate": 2.7538543328017014e-05,
|
1207 |
+
"loss": 3.7575,
|
1208 |
+
"step": 16900
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 2.7113237639553427,
|
1212 |
+
"grad_norm": 797812.75,
|
1213 |
+
"learning_rate": 2.7405635300372147e-05,
|
1214 |
+
"loss": 3.6586,
|
1215 |
+
"step": 17000
|
1216 |
+
},
|
1217 |
+
{
|
1218 |
+
"epoch": 2.7272727272727275,
|
1219 |
+
"grad_norm": 662883.0,
|
1220 |
+
"learning_rate": 2.7272727272727273e-05,
|
1221 |
+
"loss": 3.6812,
|
1222 |
+
"step": 17100
|
1223 |
+
},
|
1224 |
+
{
|
1225 |
+
"epoch": 2.7432216905901115,
|
1226 |
+
"grad_norm": 708693.8125,
|
1227 |
+
"learning_rate": 2.7139819245082403e-05,
|
1228 |
+
"loss": 3.6645,
|
1229 |
+
"step": 17200
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 2.7591706539074963,
|
1233 |
+
"grad_norm": 750630.625,
|
1234 |
+
"learning_rate": 2.7006911217437536e-05,
|
1235 |
+
"loss": 3.659,
|
1236 |
+
"step": 17300
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 2.77511961722488,
|
1240 |
+
"grad_norm": 743643.5625,
|
1241 |
+
"learning_rate": 2.6874003189792662e-05,
|
1242 |
+
"loss": 3.6794,
|
1243 |
+
"step": 17400
|
1244 |
+
},
|
1245 |
+
{
|
1246 |
+
"epoch": 2.7910685805422646,
|
1247 |
+
"grad_norm": 683095.9375,
|
1248 |
+
"learning_rate": 2.6741095162147795e-05,
|
1249 |
+
"loss": 3.6587,
|
1250 |
+
"step": 17500
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 2.807017543859649,
|
1254 |
+
"grad_norm": 659289.8125,
|
1255 |
+
"learning_rate": 2.6608187134502928e-05,
|
1256 |
+
"loss": 3.7067,
|
1257 |
+
"step": 17600
|
1258 |
+
},
|
1259 |
+
{
|
1260 |
+
"epoch": 2.8229665071770333,
|
1261 |
+
"grad_norm": 722875.9375,
|
1262 |
+
"learning_rate": 2.6475279106858054e-05,
|
1263 |
+
"loss": 3.6918,
|
1264 |
+
"step": 17700
|
1265 |
+
},
|
1266 |
+
{
|
1267 |
+
"epoch": 2.8389154704944177,
|
1268 |
+
"grad_norm": 643060.4375,
|
1269 |
+
"learning_rate": 2.6342371079213184e-05,
|
1270 |
+
"loss": 3.694,
|
1271 |
+
"step": 17800
|
1272 |
+
},
|
1273 |
+
{
|
1274 |
+
"epoch": 2.854864433811802,
|
1275 |
+
"grad_norm": 811117.3125,
|
1276 |
+
"learning_rate": 2.6209463051568317e-05,
|
1277 |
+
"loss": 3.703,
|
1278 |
+
"step": 17900
|
1279 |
+
},
|
1280 |
+
{
|
1281 |
+
"epoch": 2.8708133971291865,
|
1282 |
+
"grad_norm": 835076.6875,
|
1283 |
+
"learning_rate": 2.6076555023923443e-05,
|
1284 |
+
"loss": 3.6346,
|
1285 |
+
"step": 18000
|
1286 |
+
},
|
1287 |
+
{
|
1288 |
+
"epoch": 2.886762360446571,
|
1289 |
+
"grad_norm": 800111.875,
|
1290 |
+
"learning_rate": 2.5943646996278576e-05,
|
1291 |
+
"loss": 3.6589,
|
1292 |
+
"step": 18100
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 2.9027113237639552,
|
1296 |
+
"grad_norm": 726367.5,
|
1297 |
+
"learning_rate": 2.5810738968633706e-05,
|
1298 |
+
"loss": 3.6522,
|
1299 |
+
"step": 18200
|
1300 |
+
},
|
1301 |
+
{
|
1302 |
+
"epoch": 2.9186602870813396,
|
1303 |
+
"grad_norm": 717922.25,
|
1304 |
+
"learning_rate": 2.567783094098884e-05,
|
1305 |
+
"loss": 3.6592,
|
1306 |
+
"step": 18300
|
1307 |
+
},
|
1308 |
+
{
|
1309 |
+
"epoch": 2.934609250398724,
|
1310 |
+
"grad_norm": 841302.0625,
|
1311 |
+
"learning_rate": 2.5544922913343965e-05,
|
1312 |
+
"loss": 3.688,
|
1313 |
+
"step": 18400
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 2.9505582137161084,
|
1317 |
+
"grad_norm": 697304.1875,
|
1318 |
+
"learning_rate": 2.5412014885699098e-05,
|
1319 |
+
"loss": 3.6017,
|
1320 |
+
"step": 18500
|
1321 |
+
},
|
1322 |
+
{
|
1323 |
+
"epoch": 2.9665071770334928,
|
1324 |
+
"grad_norm": 929050.25,
|
1325 |
+
"learning_rate": 2.527910685805423e-05,
|
1326 |
+
"loss": 3.6116,
|
1327 |
+
"step": 18600
|
1328 |
+
},
|
1329 |
+
{
|
1330 |
+
"epoch": 2.982456140350877,
|
1331 |
+
"grad_norm": 699048.9375,
|
1332 |
+
"learning_rate": 2.5146198830409358e-05,
|
1333 |
+
"loss": 3.6474,
|
1334 |
+
"step": 18700
|
1335 |
+
},
|
1336 |
+
{
|
1337 |
+
"epoch": 2.9984051036682615,
|
1338 |
+
"grad_norm": 783686.0625,
|
1339 |
+
"learning_rate": 2.5013290802764487e-05,
|
1340 |
+
"loss": 3.6783,
|
1341 |
+
"step": 18800
|
1342 |
+
},
|
1343 |
+
{
|
1344 |
+
"epoch": 3.0,
|
1345 |
+
"eval_loss": 3.706315755844116,
|
1346 |
+
"eval_runtime": 212.8533,
|
1347 |
+
"eval_samples_per_second": 134.473,
|
1348 |
+
"eval_steps_per_second": 4.205,
|
1349 |
+
"step": 18810
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 3.014354066985646,
|
1353 |
+
"grad_norm": 709059.3125,
|
1354 |
+
"learning_rate": 2.4880382775119617e-05,
|
1355 |
+
"loss": 3.5423,
|
1356 |
+
"step": 18900
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 3.0303030303030303,
|
1360 |
+
"grad_norm": 655684.75,
|
1361 |
+
"learning_rate": 2.474747474747475e-05,
|
1362 |
+
"loss": 3.5606,
|
1363 |
+
"step": 19000
|
1364 |
+
},
|
1365 |
+
{
|
1366 |
+
"epoch": 3.0462519936204147,
|
1367 |
+
"grad_norm": 759915.1875,
|
1368 |
+
"learning_rate": 2.461456671982988e-05,
|
1369 |
+
"loss": 3.6979,
|
1370 |
+
"step": 19100
|
1371 |
+
},
|
1372 |
+
{
|
1373 |
+
"epoch": 3.062200956937799,
|
1374 |
+
"grad_norm": 835860.375,
|
1375 |
+
"learning_rate": 2.448165869218501e-05,
|
1376 |
+
"loss": 3.6324,
|
1377 |
+
"step": 19200
|
1378 |
+
},
|
1379 |
+
{
|
1380 |
+
"epoch": 3.0781499202551834,
|
1381 |
+
"grad_norm": 675779.5625,
|
1382 |
+
"learning_rate": 2.434875066454014e-05,
|
1383 |
+
"loss": 3.5773,
|
1384 |
+
"step": 19300
|
1385 |
+
},
|
1386 |
+
{
|
1387 |
+
"epoch": 3.094098883572568,
|
1388 |
+
"grad_norm": 700443.875,
|
1389 |
+
"learning_rate": 2.421584263689527e-05,
|
1390 |
+
"loss": 3.6309,
|
1391 |
+
"step": 19400
|
1392 |
+
},
|
1393 |
+
{
|
1394 |
+
"epoch": 3.110047846889952,
|
1395 |
+
"grad_norm": 768801.6875,
|
1396 |
+
"learning_rate": 2.4082934609250398e-05,
|
1397 |
+
"loss": 3.639,
|
1398 |
+
"step": 19500
|
1399 |
+
},
|
1400 |
+
{
|
1401 |
+
"epoch": 3.1259968102073366,
|
1402 |
+
"grad_norm": 705109.6875,
|
1403 |
+
"learning_rate": 2.395002658160553e-05,
|
1404 |
+
"loss": 3.6365,
|
1405 |
+
"step": 19600
|
1406 |
+
},
|
1407 |
+
{
|
1408 |
+
"epoch": 3.141945773524721,
|
1409 |
+
"grad_norm": 661887.5,
|
1410 |
+
"learning_rate": 2.381711855396066e-05,
|
1411 |
+
"loss": 3.5754,
|
1412 |
+
"step": 19700
|
1413 |
+
},
|
1414 |
+
{
|
1415 |
+
"epoch": 3.1578947368421053,
|
1416 |
+
"grad_norm": 691131.0,
|
1417 |
+
"learning_rate": 2.368421052631579e-05,
|
1418 |
+
"loss": 3.5888,
|
1419 |
+
"step": 19800
|
1420 |
+
},
|
1421 |
+
{
|
1422 |
+
"epoch": 3.1738437001594897,
|
1423 |
+
"grad_norm": 738841.625,
|
1424 |
+
"learning_rate": 2.355130249867092e-05,
|
1425 |
+
"loss": 3.6433,
|
1426 |
+
"step": 19900
|
1427 |
+
},
|
1428 |
+
{
|
1429 |
+
"epoch": 3.189792663476874,
|
1430 |
+
"grad_norm": 663372.1875,
|
1431 |
+
"learning_rate": 2.341839447102605e-05,
|
1432 |
+
"loss": 3.5642,
|
1433 |
+
"step": 20000
|
1434 |
+
},
|
1435 |
+
{
|
1436 |
+
"epoch": 3.2057416267942584,
|
1437 |
+
"grad_norm": 619548.75,
|
1438 |
+
"learning_rate": 2.3285486443381183e-05,
|
1439 |
+
"loss": 3.5864,
|
1440 |
+
"step": 20100
|
1441 |
+
},
|
1442 |
+
{
|
1443 |
+
"epoch": 3.221690590111643,
|
1444 |
+
"grad_norm": 773919.6875,
|
1445 |
+
"learning_rate": 2.3152578415736312e-05,
|
1446 |
+
"loss": 3.6072,
|
1447 |
+
"step": 20200
|
1448 |
+
},
|
1449 |
+
{
|
1450 |
+
"epoch": 3.237639553429027,
|
1451 |
+
"grad_norm": 728946.0625,
|
1452 |
+
"learning_rate": 2.3019670388091442e-05,
|
1453 |
+
"loss": 3.6212,
|
1454 |
+
"step": 20300
|
1455 |
+
},
|
1456 |
+
{
|
1457 |
+
"epoch": 3.2535885167464116,
|
1458 |
+
"grad_norm": 659326.9375,
|
1459 |
+
"learning_rate": 2.288676236044657e-05,
|
1460 |
+
"loss": 3.6068,
|
1461 |
+
"step": 20400
|
1462 |
+
},
|
1463 |
+
{
|
1464 |
+
"epoch": 3.269537480063796,
|
1465 |
+
"grad_norm": 759802.25,
|
1466 |
+
"learning_rate": 2.27538543328017e-05,
|
1467 |
+
"loss": 3.6783,
|
1468 |
+
"step": 20500
|
1469 |
+
},
|
1470 |
+
{
|
1471 |
+
"epoch": 3.2854864433811803,
|
1472 |
+
"grad_norm": 805141.9375,
|
1473 |
+
"learning_rate": 2.262094630515683e-05,
|
1474 |
+
"loss": 3.6441,
|
1475 |
+
"step": 20600
|
1476 |
+
},
|
1477 |
+
{
|
1478 |
+
"epoch": 3.3014354066985647,
|
1479 |
+
"grad_norm": 716521.875,
|
1480 |
+
"learning_rate": 2.2488038277511964e-05,
|
1481 |
+
"loss": 3.5748,
|
1482 |
+
"step": 20700
|
1483 |
+
},
|
1484 |
+
{
|
1485 |
+
"epoch": 3.317384370015949,
|
1486 |
+
"grad_norm": 920115.4375,
|
1487 |
+
"learning_rate": 2.2355130249867094e-05,
|
1488 |
+
"loss": 3.5985,
|
1489 |
+
"step": 20800
|
1490 |
+
},
|
1491 |
+
{
|
1492 |
+
"epoch": 3.3333333333333335,
|
1493 |
+
"grad_norm": 838658.0625,
|
1494 |
+
"learning_rate": 2.2222222222222223e-05,
|
1495 |
+
"loss": 3.564,
|
1496 |
+
"step": 20900
|
1497 |
+
},
|
1498 |
+
{
|
1499 |
+
"epoch": 3.349282296650718,
|
1500 |
+
"grad_norm": 838474.9375,
|
1501 |
+
"learning_rate": 2.2089314194577353e-05,
|
1502 |
+
"loss": 3.5875,
|
1503 |
+
"step": 21000
|
1504 |
+
},
|
1505 |
+
{
|
1506 |
+
"epoch": 3.3652312599681022,
|
1507 |
+
"grad_norm": 663992.25,
|
1508 |
+
"learning_rate": 2.1956406166932483e-05,
|
1509 |
+
"loss": 3.5681,
|
1510 |
+
"step": 21100
|
1511 |
+
},
|
1512 |
+
{
|
1513 |
+
"epoch": 3.3811802232854866,
|
1514 |
+
"grad_norm": 679489.9375,
|
1515 |
+
"learning_rate": 2.1823498139287616e-05,
|
1516 |
+
"loss": 3.5276,
|
1517 |
+
"step": 21200
|
1518 |
+
},
|
1519 |
+
{
|
1520 |
+
"epoch": 3.397129186602871,
|
1521 |
+
"grad_norm": 721495.6875,
|
1522 |
+
"learning_rate": 2.1690590111642745e-05,
|
1523 |
+
"loss": 3.5599,
|
1524 |
+
"step": 21300
|
1525 |
+
},
|
1526 |
+
{
|
1527 |
+
"epoch": 3.4130781499202554,
|
1528 |
+
"grad_norm": 667400.9375,
|
1529 |
+
"learning_rate": 2.1557682083997875e-05,
|
1530 |
+
"loss": 3.5554,
|
1531 |
+
"step": 21400
|
1532 |
+
},
|
1533 |
+
{
|
1534 |
+
"epoch": 3.4290271132376393,
|
1535 |
+
"grad_norm": 706122.0,
|
1536 |
+
"learning_rate": 2.1424774056353005e-05,
|
1537 |
+
"loss": 3.6543,
|
1538 |
+
"step": 21500
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 3.444976076555024,
|
1542 |
+
"grad_norm": 709513.125,
|
1543 |
+
"learning_rate": 2.1291866028708134e-05,
|
1544 |
+
"loss": 3.6568,
|
1545 |
+
"step": 21600
|
1546 |
+
},
|
1547 |
+
{
|
1548 |
+
"epoch": 3.460925039872408,
|
1549 |
+
"grad_norm": 652669.375,
|
1550 |
+
"learning_rate": 2.1158958001063264e-05,
|
1551 |
+
"loss": 3.6423,
|
1552 |
+
"step": 21700
|
1553 |
+
},
|
1554 |
+
{
|
1555 |
+
"epoch": 3.476874003189793,
|
1556 |
+
"grad_norm": 713371.625,
|
1557 |
+
"learning_rate": 2.1026049973418397e-05,
|
1558 |
+
"loss": 3.6102,
|
1559 |
+
"step": 21800
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 3.492822966507177,
|
1563 |
+
"grad_norm": 694536.875,
|
1564 |
+
"learning_rate": 2.0893141945773527e-05,
|
1565 |
+
"loss": 3.5488,
|
1566 |
+
"step": 21900
|
1567 |
+
},
|
1568 |
+
{
|
1569 |
+
"epoch": 3.5087719298245617,
|
1570 |
+
"grad_norm": 668059.9375,
|
1571 |
+
"learning_rate": 2.0760233918128656e-05,
|
1572 |
+
"loss": 3.6078,
|
1573 |
+
"step": 22000
|
1574 |
+
},
|
1575 |
+
{
|
1576 |
+
"epoch": 3.5247208931419456,
|
1577 |
+
"grad_norm": 731925.875,
|
1578 |
+
"learning_rate": 2.0627325890483786e-05,
|
1579 |
+
"loss": 3.5455,
|
1580 |
+
"step": 22100
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 3.5406698564593304,
|
1584 |
+
"grad_norm": 766777.9375,
|
1585 |
+
"learning_rate": 2.0494417862838915e-05,
|
1586 |
+
"loss": 3.5582,
|
1587 |
+
"step": 22200
|
1588 |
+
},
|
1589 |
+
{
|
1590 |
+
"epoch": 3.5566188197767143,
|
1591 |
+
"grad_norm": 755381.1875,
|
1592 |
+
"learning_rate": 2.0361509835194045e-05,
|
1593 |
+
"loss": 3.5859,
|
1594 |
+
"step": 22300
|
1595 |
+
},
|
1596 |
+
{
|
1597 |
+
"epoch": 3.5725677830940987,
|
1598 |
+
"grad_norm": 749873.6875,
|
1599 |
+
"learning_rate": 2.0228601807549178e-05,
|
1600 |
+
"loss": 3.7161,
|
1601 |
+
"step": 22400
|
1602 |
+
},
|
1603 |
+
{
|
1604 |
+
"epoch": 3.588516746411483,
|
1605 |
+
"grad_norm": 677539.125,
|
1606 |
+
"learning_rate": 2.0095693779904308e-05,
|
1607 |
+
"loss": 3.6225,
|
1608 |
+
"step": 22500
|
1609 |
+
},
|
1610 |
+
{
|
1611 |
+
"epoch": 3.6044657097288675,
|
1612 |
+
"grad_norm": 744990.8125,
|
1613 |
+
"learning_rate": 1.9962785752259437e-05,
|
1614 |
+
"loss": 3.5935,
|
1615 |
+
"step": 22600
|
1616 |
+
},
|
1617 |
+
{
|
1618 |
+
"epoch": 3.620414673046252,
|
1619 |
+
"grad_norm": 629363.0,
|
1620 |
+
"learning_rate": 1.9829877724614567e-05,
|
1621 |
+
"loss": 3.5828,
|
1622 |
+
"step": 22700
|
1623 |
+
},
|
1624 |
+
{
|
1625 |
+
"epoch": 3.6363636363636362,
|
1626 |
+
"grad_norm": 685327.4375,
|
1627 |
+
"learning_rate": 1.9696969696969697e-05,
|
1628 |
+
"loss": 3.5937,
|
1629 |
+
"step": 22800
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 3.6523125996810206,
|
1633 |
+
"grad_norm": 760474.25,
|
1634 |
+
"learning_rate": 1.956406166932483e-05,
|
1635 |
+
"loss": 3.5799,
|
1636 |
+
"step": 22900
|
1637 |
+
},
|
1638 |
+
{
|
1639 |
+
"epoch": 3.668261562998405,
|
1640 |
+
"grad_norm": 725888.9375,
|
1641 |
+
"learning_rate": 1.943115364167996e-05,
|
1642 |
+
"loss": 3.5465,
|
1643 |
+
"step": 23000
|
1644 |
+
},
|
1645 |
+
{
|
1646 |
+
"epoch": 3.6842105263157894,
|
1647 |
+
"grad_norm": 752312.0625,
|
1648 |
+
"learning_rate": 1.929824561403509e-05,
|
1649 |
+
"loss": 3.6185,
|
1650 |
+
"step": 23100
|
1651 |
+
},
|
1652 |
+
{
|
1653 |
+
"epoch": 3.7001594896331738,
|
1654 |
+
"grad_norm": 855498.5625,
|
1655 |
+
"learning_rate": 1.916533758639022e-05,
|
1656 |
+
"loss": 3.5792,
|
1657 |
+
"step": 23200
|
1658 |
+
},
|
1659 |
+
{
|
1660 |
+
"epoch": 3.716108452950558,
|
1661 |
+
"grad_norm": 679264.5,
|
1662 |
+
"learning_rate": 1.9032429558745348e-05,
|
1663 |
+
"loss": 3.5531,
|
1664 |
+
"step": 23300
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 3.7320574162679425,
|
1668 |
+
"grad_norm": 688482.875,
|
1669 |
+
"learning_rate": 1.8899521531100478e-05,
|
1670 |
+
"loss": 3.5495,
|
1671 |
+
"step": 23400
|
1672 |
+
},
|
1673 |
+
{
|
1674 |
+
"epoch": 3.748006379585327,
|
1675 |
+
"grad_norm": 756693.75,
|
1676 |
+
"learning_rate": 1.876661350345561e-05,
|
1677 |
+
"loss": 3.567,
|
1678 |
+
"step": 23500
|
1679 |
+
},
|
1680 |
+
{
|
1681 |
+
"epoch": 3.7639553429027113,
|
1682 |
+
"grad_norm": 854033.0625,
|
1683 |
+
"learning_rate": 1.863370547581074e-05,
|
1684 |
+
"loss": 3.5611,
|
1685 |
+
"step": 23600
|
1686 |
+
},
|
1687 |
+
{
|
1688 |
+
"epoch": 3.7799043062200957,
|
1689 |
+
"grad_norm": 753585.0625,
|
1690 |
+
"learning_rate": 1.850079744816587e-05,
|
1691 |
+
"loss": 3.5439,
|
1692 |
+
"step": 23700
|
1693 |
+
},
|
1694 |
+
{
|
1695 |
+
"epoch": 3.79585326953748,
|
1696 |
+
"grad_norm": 685358.25,
|
1697 |
+
"learning_rate": 1.8367889420521e-05,
|
1698 |
+
"loss": 3.535,
|
1699 |
+
"step": 23800
|
1700 |
+
},
|
1701 |
+
{
|
1702 |
+
"epoch": 3.8118022328548644,
|
1703 |
+
"grad_norm": 813028.4375,
|
1704 |
+
"learning_rate": 1.823498139287613e-05,
|
1705 |
+
"loss": 3.566,
|
1706 |
+
"step": 23900
|
1707 |
+
},
|
1708 |
+
{
|
1709 |
+
"epoch": 3.827751196172249,
|
1710 |
+
"grad_norm": 676295.9375,
|
1711 |
+
"learning_rate": 1.8102073365231263e-05,
|
1712 |
+
"loss": 3.5208,
|
1713 |
+
"step": 24000
|
1714 |
+
},
|
1715 |
+
{
|
1716 |
+
"epoch": 3.843700159489633,
|
1717 |
+
"grad_norm": 705614.1875,
|
1718 |
+
"learning_rate": 1.7969165337586392e-05,
|
1719 |
+
"loss": 3.5108,
|
1720 |
+
"step": 24100
|
1721 |
+
},
|
1722 |
+
{
|
1723 |
+
"epoch": 3.8596491228070176,
|
1724 |
+
"grad_norm": 643356.125,
|
1725 |
+
"learning_rate": 1.7836257309941522e-05,
|
1726 |
+
"loss": 3.5416,
|
1727 |
+
"step": 24200
|
1728 |
+
},
|
1729 |
+
{
|
1730 |
+
"epoch": 3.875598086124402,
|
1731 |
+
"grad_norm": 783635.625,
|
1732 |
+
"learning_rate": 1.770334928229665e-05,
|
1733 |
+
"loss": 3.6022,
|
1734 |
+
"step": 24300
|
1735 |
+
},
|
1736 |
+
{
|
1737 |
+
"epoch": 3.8915470494417863,
|
1738 |
+
"grad_norm": 829787.5,
|
1739 |
+
"learning_rate": 1.757044125465178e-05,
|
1740 |
+
"loss": 3.6056,
|
1741 |
+
"step": 24400
|
1742 |
+
},
|
1743 |
+
{
|
1744 |
+
"epoch": 3.9074960127591707,
|
1745 |
+
"grad_norm": 776633.0625,
|
1746 |
+
"learning_rate": 1.743753322700691e-05,
|
1747 |
+
"loss": 3.5857,
|
1748 |
+
"step": 24500
|
1749 |
+
},
|
1750 |
+
{
|
1751 |
+
"epoch": 3.923444976076555,
|
1752 |
+
"grad_norm": 708855.1875,
|
1753 |
+
"learning_rate": 1.7304625199362044e-05,
|
1754 |
+
"loss": 3.6086,
|
1755 |
+
"step": 24600
|
1756 |
+
},
|
1757 |
+
{
|
1758 |
+
"epoch": 3.9393939393939394,
|
1759 |
+
"grad_norm": 906587.0,
|
1760 |
+
"learning_rate": 1.7171717171717173e-05,
|
1761 |
+
"loss": 3.5361,
|
1762 |
+
"step": 24700
|
1763 |
+
},
|
1764 |
+
{
|
1765 |
+
"epoch": 3.955342902711324,
|
1766 |
+
"grad_norm": 835950.4375,
|
1767 |
+
"learning_rate": 1.7038809144072303e-05,
|
1768 |
+
"loss": 3.5976,
|
1769 |
+
"step": 24800
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 3.971291866028708,
|
1773 |
+
"grad_norm": 816570.6875,
|
1774 |
+
"learning_rate": 1.6905901116427433e-05,
|
1775 |
+
"loss": 3.596,
|
1776 |
+
"step": 24900
|
1777 |
+
},
|
1778 |
+
{
|
1779 |
+
"epoch": 3.9872408293460926,
|
1780 |
+
"grad_norm": 751378.9375,
|
1781 |
+
"learning_rate": 1.6772993088782562e-05,
|
1782 |
+
"loss": 3.4856,
|
1783 |
+
"step": 25000
|
1784 |
+
},
|
1785 |
+
{
|
1786 |
+
"epoch": 4.0,
|
1787 |
+
"eval_loss": 3.646216869354248,
|
1788 |
+
"eval_runtime": 212.927,
|
1789 |
+
"eval_samples_per_second": 134.426,
|
1790 |
+
"eval_steps_per_second": 4.203,
|
1791 |
+
"step": 25080
|
1792 |
+
},
|
1793 |
+
{
|
1794 |
+
"epoch": 4.003189792663477,
|
1795 |
+
"grad_norm": 735559.375,
|
1796 |
+
"learning_rate": 1.6640085061137695e-05,
|
1797 |
+
"loss": 3.5286,
|
1798 |
+
"step": 25100
|
1799 |
+
},
|
1800 |
+
{
|
1801 |
+
"epoch": 4.019138755980861,
|
1802 |
+
"grad_norm": 741852.3125,
|
1803 |
+
"learning_rate": 1.6507177033492825e-05,
|
1804 |
+
"loss": 3.5367,
|
1805 |
+
"step": 25200
|
1806 |
+
},
|
1807 |
+
{
|
1808 |
+
"epoch": 4.035087719298246,
|
1809 |
+
"grad_norm": 650216.625,
|
1810 |
+
"learning_rate": 1.6374269005847955e-05,
|
1811 |
+
"loss": 3.5056,
|
1812 |
+
"step": 25300
|
1813 |
+
},
|
1814 |
+
{
|
1815 |
+
"epoch": 4.05103668261563,
|
1816 |
+
"grad_norm": 717938.9375,
|
1817 |
+
"learning_rate": 1.6241360978203084e-05,
|
1818 |
+
"loss": 3.4996,
|
1819 |
+
"step": 25400
|
1820 |
+
},
|
1821 |
+
{
|
1822 |
+
"epoch": 4.0669856459330145,
|
1823 |
+
"grad_norm": 840426.9375,
|
1824 |
+
"learning_rate": 1.6108452950558214e-05,
|
1825 |
+
"loss": 3.5293,
|
1826 |
+
"step": 25500
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 4.082934609250398,
|
1830 |
+
"grad_norm": 699668.5,
|
1831 |
+
"learning_rate": 1.5975544922913344e-05,
|
1832 |
+
"loss": 3.5122,
|
1833 |
+
"step": 25600
|
1834 |
+
},
|
1835 |
+
{
|
1836 |
+
"epoch": 4.098883572567783,
|
1837 |
+
"grad_norm": 649238.8125,
|
1838 |
+
"learning_rate": 1.5842636895268477e-05,
|
1839 |
+
"loss": 3.4949,
|
1840 |
+
"step": 25700
|
1841 |
+
},
|
1842 |
+
{
|
1843 |
+
"epoch": 4.114832535885167,
|
1844 |
+
"grad_norm": 669479.3125,
|
1845 |
+
"learning_rate": 1.5709728867623606e-05,
|
1846 |
+
"loss": 3.5515,
|
1847 |
+
"step": 25800
|
1848 |
+
},
|
1849 |
+
{
|
1850 |
+
"epoch": 4.130781499202552,
|
1851 |
+
"grad_norm": 645182.625,
|
1852 |
+
"learning_rate": 1.5576820839978733e-05,
|
1853 |
+
"loss": 3.5113,
|
1854 |
+
"step": 25900
|
1855 |
+
},
|
1856 |
+
{
|
1857 |
+
"epoch": 4.146730462519936,
|
1858 |
+
"grad_norm": 913136.6875,
|
1859 |
+
"learning_rate": 1.5443912812333866e-05,
|
1860 |
+
"loss": 3.5326,
|
1861 |
+
"step": 26000
|
1862 |
+
},
|
1863 |
+
{
|
1864 |
+
"epoch": 4.162679425837321,
|
1865 |
+
"grad_norm": 641233.3125,
|
1866 |
+
"learning_rate": 1.5311004784688995e-05,
|
1867 |
+
"loss": 3.4768,
|
1868 |
+
"step": 26100
|
1869 |
+
},
|
1870 |
+
{
|
1871 |
+
"epoch": 4.178628389154705,
|
1872 |
+
"grad_norm": 798713.5625,
|
1873 |
+
"learning_rate": 1.5178096757044127e-05,
|
1874 |
+
"loss": 3.4977,
|
1875 |
+
"step": 26200
|
1876 |
+
},
|
1877 |
+
{
|
1878 |
+
"epoch": 4.1945773524720895,
|
1879 |
+
"grad_norm": 719145.9375,
|
1880 |
+
"learning_rate": 1.5045188729399256e-05,
|
1881 |
+
"loss": 3.4549,
|
1882 |
+
"step": 26300
|
1883 |
+
},
|
1884 |
+
{
|
1885 |
+
"epoch": 4.2105263157894735,
|
1886 |
+
"grad_norm": 692601.9375,
|
1887 |
+
"learning_rate": 1.4912280701754386e-05,
|
1888 |
+
"loss": 3.5228,
|
1889 |
+
"step": 26400
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 4.226475279106858,
|
1893 |
+
"grad_norm": 631408.4375,
|
1894 |
+
"learning_rate": 1.4779372674109517e-05,
|
1895 |
+
"loss": 3.4742,
|
1896 |
+
"step": 26500
|
1897 |
+
},
|
1898 |
+
{
|
1899 |
+
"epoch": 4.242424242424242,
|
1900 |
+
"grad_norm": 733137.0,
|
1901 |
+
"learning_rate": 1.4646464646464647e-05,
|
1902 |
+
"loss": 3.4242,
|
1903 |
+
"step": 26600
|
1904 |
+
},
|
1905 |
+
{
|
1906 |
+
"epoch": 4.258373205741627,
|
1907 |
+
"grad_norm": 741812.9375,
|
1908 |
+
"learning_rate": 1.4513556618819777e-05,
|
1909 |
+
"loss": 3.4598,
|
1910 |
+
"step": 26700
|
1911 |
+
},
|
1912 |
+
{
|
1913 |
+
"epoch": 4.274322169059011,
|
1914 |
+
"grad_norm": 682720.125,
|
1915 |
+
"learning_rate": 1.4380648591174908e-05,
|
1916 |
+
"loss": 3.4827,
|
1917 |
+
"step": 26800
|
1918 |
+
},
|
1919 |
+
{
|
1920 |
+
"epoch": 4.290271132376396,
|
1921 |
+
"grad_norm": 784646.125,
|
1922 |
+
"learning_rate": 1.4247740563530037e-05,
|
1923 |
+
"loss": 3.4712,
|
1924 |
+
"step": 26900
|
1925 |
+
},
|
1926 |
+
{
|
1927 |
+
"epoch": 4.30622009569378,
|
1928 |
+
"grad_norm": 713282.75,
|
1929 |
+
"learning_rate": 1.4114832535885167e-05,
|
1930 |
+
"loss": 3.5542,
|
1931 |
+
"step": 27000
|
1932 |
+
},
|
1933 |
+
{
|
1934 |
+
"epoch": 4.3221690590111645,
|
1935 |
+
"grad_norm": 699440.4375,
|
1936 |
+
"learning_rate": 1.3981924508240298e-05,
|
1937 |
+
"loss": 3.4572,
|
1938 |
+
"step": 27100
|
1939 |
+
},
|
1940 |
+
{
|
1941 |
+
"epoch": 4.3381180223285485,
|
1942 |
+
"grad_norm": 733063.875,
|
1943 |
+
"learning_rate": 1.3849016480595428e-05,
|
1944 |
+
"loss": 3.4106,
|
1945 |
+
"step": 27200
|
1946 |
+
},
|
1947 |
+
{
|
1948 |
+
"epoch": 4.354066985645933,
|
1949 |
+
"grad_norm": 799606.875,
|
1950 |
+
"learning_rate": 1.371610845295056e-05,
|
1951 |
+
"loss": 3.4189,
|
1952 |
+
"step": 27300
|
1953 |
+
},
|
1954 |
+
{
|
1955 |
+
"epoch": 4.370015948963317,
|
1956 |
+
"grad_norm": 722583.25,
|
1957 |
+
"learning_rate": 1.3583200425305689e-05,
|
1958 |
+
"loss": 3.448,
|
1959 |
+
"step": 27400
|
1960 |
+
},
|
1961 |
+
{
|
1962 |
+
"epoch": 4.385964912280702,
|
1963 |
+
"grad_norm": 761491.5625,
|
1964 |
+
"learning_rate": 1.3450292397660819e-05,
|
1965 |
+
"loss": 3.5248,
|
1966 |
+
"step": 27500
|
1967 |
+
},
|
1968 |
+
{
|
1969 |
+
"epoch": 4.401913875598086,
|
1970 |
+
"grad_norm": 633397.6875,
|
1971 |
+
"learning_rate": 1.331738437001595e-05,
|
1972 |
+
"loss": 3.5675,
|
1973 |
+
"step": 27600
|
1974 |
+
},
|
1975 |
+
{
|
1976 |
+
"epoch": 4.417862838915471,
|
1977 |
+
"grad_norm": 743160.4375,
|
1978 |
+
"learning_rate": 1.318447634237108e-05,
|
1979 |
+
"loss": 3.4983,
|
1980 |
+
"step": 27700
|
1981 |
+
},
|
1982 |
+
{
|
1983 |
+
"epoch": 4.433811802232855,
|
1984 |
+
"grad_norm": 689363.3125,
|
1985 |
+
"learning_rate": 1.305156831472621e-05,
|
1986 |
+
"loss": 3.5689,
|
1987 |
+
"step": 27800
|
1988 |
+
},
|
1989 |
+
{
|
1990 |
+
"epoch": 4.44976076555024,
|
1991 |
+
"grad_norm": 634674.0,
|
1992 |
+
"learning_rate": 1.291866028708134e-05,
|
1993 |
+
"loss": 3.4763,
|
1994 |
+
"step": 27900
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 4.4657097288676235,
|
1998 |
+
"grad_norm": 682868.5,
|
1999 |
+
"learning_rate": 1.278575225943647e-05,
|
2000 |
+
"loss": 3.4725,
|
2001 |
+
"step": 28000
|
2002 |
+
},
|
2003 |
+
{
|
2004 |
+
"epoch": 4.481658692185008,
|
2005 |
+
"grad_norm": 839863.0,
|
2006 |
+
"learning_rate": 1.26528442317916e-05,
|
2007 |
+
"loss": 3.5255,
|
2008 |
+
"step": 28100
|
2009 |
+
},
|
2010 |
+
{
|
2011 |
+
"epoch": 4.497607655502392,
|
2012 |
+
"grad_norm": 840497.0625,
|
2013 |
+
"learning_rate": 1.2519936204146731e-05,
|
2014 |
+
"loss": 3.4914,
|
2015 |
+
"step": 28200
|
2016 |
+
},
|
2017 |
+
{
|
2018 |
+
"epoch": 4.513556618819777,
|
2019 |
+
"grad_norm": 798480.8125,
|
2020 |
+
"learning_rate": 1.2387028176501861e-05,
|
2021 |
+
"loss": 3.4787,
|
2022 |
+
"step": 28300
|
2023 |
+
},
|
2024 |
+
{
|
2025 |
+
"epoch": 4.529505582137161,
|
2026 |
+
"grad_norm": 776783.625,
|
2027 |
+
"learning_rate": 1.2254120148856992e-05,
|
2028 |
+
"loss": 3.4817,
|
2029 |
+
"step": 28400
|
2030 |
+
},
|
2031 |
+
{
|
2032 |
+
"epoch": 4.545454545454545,
|
2033 |
+
"grad_norm": 860058.625,
|
2034 |
+
"learning_rate": 1.2121212121212122e-05,
|
2035 |
+
"loss": 3.515,
|
2036 |
+
"step": 28500
|
2037 |
+
},
|
2038 |
+
{
|
2039 |
+
"epoch": 4.56140350877193,
|
2040 |
+
"grad_norm": 797709.0,
|
2041 |
+
"learning_rate": 1.1988304093567252e-05,
|
2042 |
+
"loss": 3.4715,
|
2043 |
+
"step": 28600
|
2044 |
+
},
|
2045 |
+
{
|
2046 |
+
"epoch": 4.577352472089315,
|
2047 |
+
"grad_norm": 686232.0,
|
2048 |
+
"learning_rate": 1.1855396065922381e-05,
|
2049 |
+
"loss": 3.493,
|
2050 |
+
"step": 28700
|
2051 |
+
},
|
2052 |
+
{
|
2053 |
+
"epoch": 4.5933014354066986,
|
2054 |
+
"grad_norm": 713671.875,
|
2055 |
+
"learning_rate": 1.1722488038277513e-05,
|
2056 |
+
"loss": 3.4767,
|
2057 |
+
"step": 28800
|
2058 |
+
},
|
2059 |
+
{
|
2060 |
+
"epoch": 4.6092503987240825,
|
2061 |
+
"grad_norm": 830858.875,
|
2062 |
+
"learning_rate": 1.1589580010632644e-05,
|
2063 |
+
"loss": 3.5506,
|
2064 |
+
"step": 28900
|
2065 |
+
},
|
2066 |
+
{
|
2067 |
+
"epoch": 4.625199362041467,
|
2068 |
+
"grad_norm": 681684.8125,
|
2069 |
+
"learning_rate": 1.1456671982987772e-05,
|
2070 |
+
"loss": 3.468,
|
2071 |
+
"step": 29000
|
2072 |
+
},
|
2073 |
+
{
|
2074 |
+
"epoch": 4.641148325358852,
|
2075 |
+
"grad_norm": 693863.25,
|
2076 |
+
"learning_rate": 1.1323763955342903e-05,
|
2077 |
+
"loss": 3.4847,
|
2078 |
+
"step": 29100
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 4.657097288676236,
|
2082 |
+
"grad_norm": 612233.75,
|
2083 |
+
"learning_rate": 1.1190855927698035e-05,
|
2084 |
+
"loss": 3.3994,
|
2085 |
+
"step": 29200
|
2086 |
+
},
|
2087 |
+
{
|
2088 |
+
"epoch": 4.67304625199362,
|
2089 |
+
"grad_norm": 901251.25,
|
2090 |
+
"learning_rate": 1.1057947900053164e-05,
|
2091 |
+
"loss": 3.5144,
|
2092 |
+
"step": 29300
|
2093 |
+
},
|
2094 |
+
{
|
2095 |
+
"epoch": 4.688995215311005,
|
2096 |
+
"grad_norm": 742618.4375,
|
2097 |
+
"learning_rate": 1.0925039872408294e-05,
|
2098 |
+
"loss": 3.5388,
|
2099 |
+
"step": 29400
|
2100 |
+
},
|
2101 |
+
{
|
2102 |
+
"epoch": 4.70494417862839,
|
2103 |
+
"grad_norm": 797654.0,
|
2104 |
+
"learning_rate": 1.0792131844763423e-05,
|
2105 |
+
"loss": 3.4722,
|
2106 |
+
"step": 29500
|
2107 |
+
},
|
2108 |
+
{
|
2109 |
+
"epoch": 4.720893141945774,
|
2110 |
+
"grad_norm": 763737.9375,
|
2111 |
+
"learning_rate": 1.0659223817118555e-05,
|
2112 |
+
"loss": 3.5006,
|
2113 |
+
"step": 29600
|
2114 |
+
},
|
2115 |
+
{
|
2116 |
+
"epoch": 4.7368421052631575,
|
2117 |
+
"grad_norm": 702353.875,
|
2118 |
+
"learning_rate": 1.0526315789473684e-05,
|
2119 |
+
"loss": 3.4772,
|
2120 |
+
"step": 29700
|
2121 |
+
},
|
2122 |
+
{
|
2123 |
+
"epoch": 4.752791068580542,
|
2124 |
+
"grad_norm": 790482.1875,
|
2125 |
+
"learning_rate": 1.0393407761828814e-05,
|
2126 |
+
"loss": 3.4507,
|
2127 |
+
"step": 29800
|
2128 |
+
},
|
2129 |
+
{
|
2130 |
+
"epoch": 4.768740031897926,
|
2131 |
+
"grad_norm": 706455.9375,
|
2132 |
+
"learning_rate": 1.0260499734183945e-05,
|
2133 |
+
"loss": 3.4769,
|
2134 |
+
"step": 29900
|
2135 |
+
},
|
2136 |
+
{
|
2137 |
+
"epoch": 4.784688995215311,
|
2138 |
+
"grad_norm": 720554.3125,
|
2139 |
+
"learning_rate": 1.0127591706539077e-05,
|
2140 |
+
"loss": 3.4551,
|
2141 |
+
"step": 30000
|
2142 |
+
},
|
2143 |
+
{
|
2144 |
+
"epoch": 4.800637958532695,
|
2145 |
+
"grad_norm": 754827.6875,
|
2146 |
+
"learning_rate": 9.994683678894205e-06,
|
2147 |
+
"loss": 3.4982,
|
2148 |
+
"step": 30100
|
2149 |
+
},
|
2150 |
+
{
|
2151 |
+
"epoch": 4.81658692185008,
|
2152 |
+
"grad_norm": 696089.375,
|
2153 |
+
"learning_rate": 9.861775651249336e-06,
|
2154 |
+
"loss": 3.4637,
|
2155 |
+
"step": 30200
|
2156 |
+
},
|
2157 |
+
{
|
2158 |
+
"epoch": 4.832535885167464,
|
2159 |
+
"grad_norm": 737095.3125,
|
2160 |
+
"learning_rate": 9.728867623604466e-06,
|
2161 |
+
"loss": 3.4621,
|
2162 |
+
"step": 30300
|
2163 |
+
},
|
2164 |
+
{
|
2165 |
+
"epoch": 4.848484848484849,
|
2166 |
+
"grad_norm": 768262.125,
|
2167 |
+
"learning_rate": 9.595959595959595e-06,
|
2168 |
+
"loss": 3.48,
|
2169 |
+
"step": 30400
|
2170 |
+
},
|
2171 |
+
{
|
2172 |
+
"epoch": 4.8644338118022326,
|
2173 |
+
"grad_norm": 767420.5625,
|
2174 |
+
"learning_rate": 9.463051568314727e-06,
|
2175 |
+
"loss": 3.4463,
|
2176 |
+
"step": 30500
|
2177 |
+
},
|
2178 |
+
{
|
2179 |
+
"epoch": 4.880382775119617,
|
2180 |
+
"grad_norm": 706310.0,
|
2181 |
+
"learning_rate": 9.330143540669856e-06,
|
2182 |
+
"loss": 3.4668,
|
2183 |
+
"step": 30600
|
2184 |
+
},
|
2185 |
+
{
|
2186 |
+
"epoch": 4.896331738437001,
|
2187 |
+
"grad_norm": 828940.625,
|
2188 |
+
"learning_rate": 9.197235513024988e-06,
|
2189 |
+
"loss": 3.4377,
|
2190 |
+
"step": 30700
|
2191 |
+
},
|
2192 |
+
{
|
2193 |
+
"epoch": 4.912280701754386,
|
2194 |
+
"grad_norm": 788075.375,
|
2195 |
+
"learning_rate": 9.064327485380117e-06,
|
2196 |
+
"loss": 3.4144,
|
2197 |
+
"step": 30800
|
2198 |
+
},
|
2199 |
+
{
|
2200 |
+
"epoch": 4.92822966507177,
|
2201 |
+
"grad_norm": 805364.0625,
|
2202 |
+
"learning_rate": 8.931419457735247e-06,
|
2203 |
+
"loss": 3.4726,
|
2204 |
+
"step": 30900
|
2205 |
+
},
|
2206 |
+
{
|
2207 |
+
"epoch": 4.944178628389155,
|
2208 |
+
"grad_norm": 604472.75,
|
2209 |
+
"learning_rate": 8.798511430090378e-06,
|
2210 |
+
"loss": 3.4936,
|
2211 |
+
"step": 31000
|
2212 |
+
},
|
2213 |
+
{
|
2214 |
+
"epoch": 4.960127591706539,
|
2215 |
+
"grad_norm": 684835.3125,
|
2216 |
+
"learning_rate": 8.66560340244551e-06,
|
2217 |
+
"loss": 3.4637,
|
2218 |
+
"step": 31100
|
2219 |
+
},
|
2220 |
+
{
|
2221 |
+
"epoch": 4.976076555023924,
|
2222 |
+
"grad_norm": 684998.5,
|
2223 |
+
"learning_rate": 8.532695374800638e-06,
|
2224 |
+
"loss": 3.5321,
|
2225 |
+
"step": 31200
|
2226 |
+
},
|
2227 |
+
{
|
2228 |
+
"epoch": 4.992025518341308,
|
2229 |
+
"grad_norm": 711871.5625,
|
2230 |
+
"learning_rate": 8.399787347155769e-06,
|
2231 |
+
"loss": 3.4809,
|
2232 |
+
"step": 31300
|
2233 |
+
},
|
2234 |
+
{
|
2235 |
+
"epoch": 5.0,
|
2236 |
+
"eval_loss": 3.6019129753112793,
|
2237 |
+
"eval_runtime": 212.6906,
|
2238 |
+
"eval_samples_per_second": 134.576,
|
2239 |
+
"eval_steps_per_second": 4.208,
|
2240 |
+
"step": 31350
|
2241 |
+
},
|
2242 |
+
{
|
2243 |
+
"epoch": 5.007974481658692,
|
2244 |
+
"grad_norm": 1041463.125,
|
2245 |
+
"learning_rate": 8.266879319510899e-06,
|
2246 |
+
"loss": 3.4055,
|
2247 |
+
"step": 31400
|
2248 |
+
},
|
2249 |
+
{
|
2250 |
+
"epoch": 5.023923444976076,
|
2251 |
+
"grad_norm": 623676.3125,
|
2252 |
+
"learning_rate": 8.133971291866028e-06,
|
2253 |
+
"loss": 3.5315,
|
2254 |
+
"step": 31500
|
2255 |
+
},
|
2256 |
+
{
|
2257 |
+
"epoch": 5.039872408293461,
|
2258 |
+
"grad_norm": 619523.9375,
|
2259 |
+
"learning_rate": 8.00106326422116e-06,
|
2260 |
+
"loss": 3.4709,
|
2261 |
+
"step": 31600
|
2262 |
+
},
|
2263 |
+
{
|
2264 |
+
"epoch": 5.055821371610845,
|
2265 |
+
"grad_norm": 779379.875,
|
2266 |
+
"learning_rate": 7.86815523657629e-06,
|
2267 |
+
"loss": 3.3933,
|
2268 |
+
"step": 31700
|
2269 |
+
},
|
2270 |
+
{
|
2271 |
+
"epoch": 5.07177033492823,
|
2272 |
+
"grad_norm": 617667.5625,
|
2273 |
+
"learning_rate": 7.73524720893142e-06,
|
2274 |
+
"loss": 3.4738,
|
2275 |
+
"step": 31800
|
2276 |
+
},
|
2277 |
+
{
|
2278 |
+
"epoch": 5.087719298245614,
|
2279 |
+
"grad_norm": 691351.5625,
|
2280 |
+
"learning_rate": 7.602339181286549e-06,
|
2281 |
+
"loss": 3.4499,
|
2282 |
+
"step": 31900
|
2283 |
+
},
|
2284 |
+
{
|
2285 |
+
"epoch": 5.103668261562999,
|
2286 |
+
"grad_norm": 702913.25,
|
2287 |
+
"learning_rate": 7.469431153641681e-06,
|
2288 |
+
"loss": 3.4476,
|
2289 |
+
"step": 32000
|
2290 |
+
},
|
2291 |
+
{
|
2292 |
+
"epoch": 5.119617224880383,
|
2293 |
+
"grad_norm": 778967.4375,
|
2294 |
+
"learning_rate": 7.336523125996811e-06,
|
2295 |
+
"loss": 3.4689,
|
2296 |
+
"step": 32100
|
2297 |
+
},
|
2298 |
+
{
|
2299 |
+
"epoch": 5.1355661881977674,
|
2300 |
+
"grad_norm": 693687.3125,
|
2301 |
+
"learning_rate": 7.20361509835194e-06,
|
2302 |
+
"loss": 3.381,
|
2303 |
+
"step": 32200
|
2304 |
+
},
|
2305 |
+
{
|
2306 |
+
"epoch": 5.151515151515151,
|
2307 |
+
"grad_norm": 843375.5,
|
2308 |
+
"learning_rate": 7.0707070707070704e-06,
|
2309 |
+
"loss": 3.438,
|
2310 |
+
"step": 32300
|
2311 |
+
},
|
2312 |
+
{
|
2313 |
+
"epoch": 5.167464114832536,
|
2314 |
+
"grad_norm": 734730.5,
|
2315 |
+
"learning_rate": 6.937799043062202e-06,
|
2316 |
+
"loss": 3.4564,
|
2317 |
+
"step": 32400
|
2318 |
+
},
|
2319 |
+
{
|
2320 |
+
"epoch": 5.18341307814992,
|
2321 |
+
"grad_norm": 716158.375,
|
2322 |
+
"learning_rate": 6.804891015417332e-06,
|
2323 |
+
"loss": 3.4529,
|
2324 |
+
"step": 32500
|
2325 |
+
},
|
2326 |
+
{
|
2327 |
+
"epoch": 5.199362041467305,
|
2328 |
+
"grad_norm": 703040.625,
|
2329 |
+
"learning_rate": 6.671982987772461e-06,
|
2330 |
+
"loss": 3.391,
|
2331 |
+
"step": 32600
|
2332 |
+
},
|
2333 |
+
{
|
2334 |
+
"epoch": 5.215311004784689,
|
2335 |
+
"grad_norm": 636837.3125,
|
2336 |
+
"learning_rate": 6.5390749601275915e-06,
|
2337 |
+
"loss": 3.4174,
|
2338 |
+
"step": 32700
|
2339 |
+
},
|
2340 |
+
{
|
2341 |
+
"epoch": 5.231259968102074,
|
2342 |
+
"grad_norm": 661893.6875,
|
2343 |
+
"learning_rate": 6.406166932482723e-06,
|
2344 |
+
"loss": 3.4256,
|
2345 |
+
"step": 32800
|
2346 |
+
},
|
2347 |
+
{
|
2348 |
+
"epoch": 5.247208931419458,
|
2349 |
+
"grad_norm": 812641.0625,
|
2350 |
+
"learning_rate": 6.273258904837853e-06,
|
2351 |
+
"loss": 3.4981,
|
2352 |
+
"step": 32900
|
2353 |
+
},
|
2354 |
+
{
|
2355 |
+
"epoch": 5.2631578947368425,
|
2356 |
+
"grad_norm": 725753.125,
|
2357 |
+
"learning_rate": 6.140350877192982e-06,
|
2358 |
+
"loss": 3.5018,
|
2359 |
+
"step": 33000
|
2360 |
+
},
|
2361 |
+
{
|
2362 |
+
"epoch": 5.279106858054226,
|
2363 |
+
"grad_norm": 691768.3125,
|
2364 |
+
"learning_rate": 6.007442849548113e-06,
|
2365 |
+
"loss": 3.3969,
|
2366 |
+
"step": 33100
|
2367 |
+
},
|
2368 |
+
{
|
2369 |
+
"epoch": 5.295055821371611,
|
2370 |
+
"grad_norm": 776131.9375,
|
2371 |
+
"learning_rate": 5.874534821903243e-06,
|
2372 |
+
"loss": 3.4489,
|
2373 |
+
"step": 33200
|
2374 |
+
},
|
2375 |
+
{
|
2376 |
+
"epoch": 5.311004784688995,
|
2377 |
+
"grad_norm": 747268.25,
|
2378 |
+
"learning_rate": 5.741626794258374e-06,
|
2379 |
+
"loss": 3.4075,
|
2380 |
+
"step": 33300
|
2381 |
+
},
|
2382 |
+
{
|
2383 |
+
"epoch": 5.32695374800638,
|
2384 |
+
"grad_norm": 864468.1875,
|
2385 |
+
"learning_rate": 5.608718766613503e-06,
|
2386 |
+
"loss": 3.4443,
|
2387 |
+
"step": 33400
|
2388 |
+
},
|
2389 |
+
{
|
2390 |
+
"epoch": 5.342902711323764,
|
2391 |
+
"grad_norm": 729350.5,
|
2392 |
+
"learning_rate": 5.475810738968634e-06,
|
2393 |
+
"loss": 3.4189,
|
2394 |
+
"step": 33500
|
2395 |
+
},
|
2396 |
+
{
|
2397 |
+
"epoch": 5.358851674641148,
|
2398 |
+
"grad_norm": 687491.0625,
|
2399 |
+
"learning_rate": 5.342902711323764e-06,
|
2400 |
+
"loss": 3.4167,
|
2401 |
+
"step": 33600
|
2402 |
+
},
|
2403 |
+
{
|
2404 |
+
"epoch": 5.374800637958533,
|
2405 |
+
"grad_norm": 682089.125,
|
2406 |
+
"learning_rate": 5.209994683678895e-06,
|
2407 |
+
"loss": 3.3691,
|
2408 |
+
"step": 33700
|
2409 |
+
},
|
2410 |
+
{
|
2411 |
+
"epoch": 5.3907496012759175,
|
2412 |
+
"grad_norm": 698959.25,
|
2413 |
+
"learning_rate": 5.077086656034024e-06,
|
2414 |
+
"loss": 3.4449,
|
2415 |
+
"step": 33800
|
2416 |
+
},
|
2417 |
+
{
|
2418 |
+
"epoch": 5.4066985645933014,
|
2419 |
+
"grad_norm": 671919.375,
|
2420 |
+
"learning_rate": 4.944178628389155e-06,
|
2421 |
+
"loss": 3.4421,
|
2422 |
+
"step": 33900
|
2423 |
+
},
|
2424 |
+
{
|
2425 |
+
"epoch": 5.422647527910685,
|
2426 |
+
"grad_norm": 647332.125,
|
2427 |
+
"learning_rate": 4.811270600744285e-06,
|
2428 |
+
"loss": 3.4529,
|
2429 |
+
"step": 34000
|
2430 |
+
},
|
2431 |
+
{
|
2432 |
+
"epoch": 5.43859649122807,
|
2433 |
+
"grad_norm": 686242.25,
|
2434 |
+
"learning_rate": 4.678362573099415e-06,
|
2435 |
+
"loss": 3.3381,
|
2436 |
+
"step": 34100
|
2437 |
+
},
|
2438 |
+
{
|
2439 |
+
"epoch": 5.454545454545454,
|
2440 |
+
"grad_norm": 936387.5,
|
2441 |
+
"learning_rate": 4.5454545454545455e-06,
|
2442 |
+
"loss": 3.4765,
|
2443 |
+
"step": 34200
|
2444 |
+
},
|
2445 |
+
{
|
2446 |
+
"epoch": 5.470494417862839,
|
2447 |
+
"grad_norm": 662910.3125,
|
2448 |
+
"learning_rate": 4.412546517809676e-06,
|
2449 |
+
"loss": 3.3729,
|
2450 |
+
"step": 34300
|
2451 |
+
},
|
2452 |
+
{
|
2453 |
+
"epoch": 5.486443381180223,
|
2454 |
+
"grad_norm": 671547.0,
|
2455 |
+
"learning_rate": 4.2796384901648065e-06,
|
2456 |
+
"loss": 3.4489,
|
2457 |
+
"step": 34400
|
2458 |
+
},
|
2459 |
+
{
|
2460 |
+
"epoch": 5.502392344497608,
|
2461 |
+
"grad_norm": 777179.3125,
|
2462 |
+
"learning_rate": 4.146730462519936e-06,
|
2463 |
+
"loss": 3.424,
|
2464 |
+
"step": 34500
|
2465 |
+
},
|
2466 |
+
{
|
2467 |
+
"epoch": 5.518341307814992,
|
2468 |
+
"grad_norm": 628742.25,
|
2469 |
+
"learning_rate": 4.013822434875067e-06,
|
2470 |
+
"loss": 3.4128,
|
2471 |
+
"step": 34600
|
2472 |
+
},
|
2473 |
+
{
|
2474 |
+
"epoch": 5.5342902711323765,
|
2475 |
+
"grad_norm": 794101.75,
|
2476 |
+
"learning_rate": 3.880914407230197e-06,
|
2477 |
+
"loss": 3.3938,
|
2478 |
+
"step": 34700
|
2479 |
+
},
|
2480 |
+
{
|
2481 |
+
"epoch": 5.55023923444976,
|
2482 |
+
"grad_norm": 720820.4375,
|
2483 |
+
"learning_rate": 3.7480063795853268e-06,
|
2484 |
+
"loss": 3.3622,
|
2485 |
+
"step": 34800
|
2486 |
+
},
|
2487 |
+
{
|
2488 |
+
"epoch": 5.566188197767145,
|
2489 |
+
"grad_norm": 686912.5,
|
2490 |
+
"learning_rate": 3.6150983519404573e-06,
|
2491 |
+
"loss": 3.3896,
|
2492 |
+
"step": 34900
|
2493 |
+
},
|
2494 |
+
{
|
2495 |
+
"epoch": 5.582137161084529,
|
2496 |
+
"grad_norm": 758477.0,
|
2497 |
+
"learning_rate": 3.4821903242955873e-06,
|
2498 |
+
"loss": 3.4494,
|
2499 |
+
"step": 35000
|
2500 |
+
},
|
2501 |
+
{
|
2502 |
+
"epoch": 5.598086124401914,
|
2503 |
+
"grad_norm": 828910.875,
|
2504 |
+
"learning_rate": 3.349282296650718e-06,
|
2505 |
+
"loss": 3.4204,
|
2506 |
+
"step": 35100
|
2507 |
+
},
|
2508 |
+
{
|
2509 |
+
"epoch": 5.614035087719298,
|
2510 |
+
"grad_norm": 650998.8125,
|
2511 |
+
"learning_rate": 3.216374269005848e-06,
|
2512 |
+
"loss": 3.4115,
|
2513 |
+
"step": 35200
|
2514 |
+
},
|
2515 |
+
{
|
2516 |
+
"epoch": 5.629984051036683,
|
2517 |
+
"grad_norm": 681491.375,
|
2518 |
+
"learning_rate": 3.0834662413609784e-06,
|
2519 |
+
"loss": 3.3861,
|
2520 |
+
"step": 35300
|
2521 |
+
},
|
2522 |
+
{
|
2523 |
+
"epoch": 5.645933014354067,
|
2524 |
+
"grad_norm": 749556.875,
|
2525 |
+
"learning_rate": 2.9505582137161084e-06,
|
2526 |
+
"loss": 3.4175,
|
2527 |
+
"step": 35400
|
2528 |
+
},
|
2529 |
+
{
|
2530 |
+
"epoch": 5.6618819776714515,
|
2531 |
+
"grad_norm": 677371.6875,
|
2532 |
+
"learning_rate": 2.817650186071239e-06,
|
2533 |
+
"loss": 3.4571,
|
2534 |
+
"step": 35500
|
2535 |
+
},
|
2536 |
+
{
|
2537 |
+
"epoch": 5.6778309409888355,
|
2538 |
+
"grad_norm": 730101.5625,
|
2539 |
+
"learning_rate": 2.684742158426369e-06,
|
2540 |
+
"loss": 3.3966,
|
2541 |
+
"step": 35600
|
2542 |
+
},
|
2543 |
+
{
|
2544 |
+
"epoch": 5.69377990430622,
|
2545 |
+
"grad_norm": 657353.375,
|
2546 |
+
"learning_rate": 2.5518341307814995e-06,
|
2547 |
+
"loss": 3.4639,
|
2548 |
+
"step": 35700
|
2549 |
+
},
|
2550 |
+
{
|
2551 |
+
"epoch": 5.709728867623604,
|
2552 |
+
"grad_norm": 734038.0,
|
2553 |
+
"learning_rate": 2.4189261031366296e-06,
|
2554 |
+
"loss": 3.3982,
|
2555 |
+
"step": 35800
|
2556 |
+
},
|
2557 |
+
{
|
2558 |
+
"epoch": 5.725677830940989,
|
2559 |
+
"grad_norm": 719495.625,
|
2560 |
+
"learning_rate": 2.28601807549176e-06,
|
2561 |
+
"loss": 3.4221,
|
2562 |
+
"step": 35900
|
2563 |
+
},
|
2564 |
+
{
|
2565 |
+
"epoch": 5.741626794258373,
|
2566 |
+
"grad_norm": 686361.0625,
|
2567 |
+
"learning_rate": 2.15311004784689e-06,
|
2568 |
+
"loss": 3.4886,
|
2569 |
+
"step": 36000
|
2570 |
+
},
|
2571 |
+
{
|
2572 |
+
"epoch": 5.757575757575758,
|
2573 |
+
"grad_norm": 804130.625,
|
2574 |
+
"learning_rate": 2.0202020202020206e-06,
|
2575 |
+
"loss": 3.4364,
|
2576 |
+
"step": 36100
|
2577 |
+
},
|
2578 |
+
{
|
2579 |
+
"epoch": 5.773524720893142,
|
2580 |
+
"grad_norm": 713737.5625,
|
2581 |
+
"learning_rate": 1.8872939925571505e-06,
|
2582 |
+
"loss": 3.391,
|
2583 |
+
"step": 36200
|
2584 |
+
},
|
2585 |
+
{
|
2586 |
+
"epoch": 5.7894736842105265,
|
2587 |
+
"grad_norm": 734503.125,
|
2588 |
+
"learning_rate": 1.7543859649122807e-06,
|
2589 |
+
"loss": 3.4291,
|
2590 |
+
"step": 36300
|
2591 |
+
},
|
2592 |
+
{
|
2593 |
+
"epoch": 5.8054226475279105,
|
2594 |
+
"grad_norm": 673341.3125,
|
2595 |
+
"learning_rate": 1.621477937267411e-06,
|
2596 |
+
"loss": 3.4816,
|
2597 |
+
"step": 36400
|
2598 |
+
},
|
2599 |
+
{
|
2600 |
+
"epoch": 5.821371610845295,
|
2601 |
+
"grad_norm": 643799.3125,
|
2602 |
+
"learning_rate": 1.4885699096225413e-06,
|
2603 |
+
"loss": 3.4221,
|
2604 |
+
"step": 36500
|
2605 |
+
},
|
2606 |
+
{
|
2607 |
+
"epoch": 5.837320574162679,
|
2608 |
+
"grad_norm": 641417.125,
|
2609 |
+
"learning_rate": 1.3556618819776716e-06,
|
2610 |
+
"loss": 3.4331,
|
2611 |
+
"step": 36600
|
2612 |
+
},
|
2613 |
+
{
|
2614 |
+
"epoch": 5.853269537480064,
|
2615 |
+
"grad_norm": 707790.25,
|
2616 |
+
"learning_rate": 1.2227538543328019e-06,
|
2617 |
+
"loss": 3.3936,
|
2618 |
+
"step": 36700
|
2619 |
+
},
|
2620 |
+
{
|
2621 |
+
"epoch": 5.869218500797448,
|
2622 |
+
"grad_norm": 734012.3125,
|
2623 |
+
"learning_rate": 1.089845826687932e-06,
|
2624 |
+
"loss": 3.4258,
|
2625 |
+
"step": 36800
|
2626 |
+
},
|
2627 |
+
{
|
2628 |
+
"epoch": 5.885167464114833,
|
2629 |
+
"grad_norm": 672716.0,
|
2630 |
+
"learning_rate": 9.569377990430622e-07,
|
2631 |
+
"loss": 3.4688,
|
2632 |
+
"step": 36900
|
2633 |
+
},
|
2634 |
+
{
|
2635 |
+
"epoch": 5.901116427432217,
|
2636 |
+
"grad_norm": 691052.75,
|
2637 |
+
"learning_rate": 8.240297713981925e-07,
|
2638 |
+
"loss": 3.3553,
|
2639 |
+
"step": 37000
|
2640 |
+
},
|
2641 |
+
{
|
2642 |
+
"epoch": 5.917065390749602,
|
2643 |
+
"grad_norm": 714107.875,
|
2644 |
+
"learning_rate": 6.911217437533228e-07,
|
2645 |
+
"loss": 3.4735,
|
2646 |
+
"step": 37100
|
2647 |
+
},
|
2648 |
+
{
|
2649 |
+
"epoch": 5.9330143540669855,
|
2650 |
+
"grad_norm": 783142.9375,
|
2651 |
+
"learning_rate": 5.582137161084529e-07,
|
2652 |
+
"loss": 3.4707,
|
2653 |
+
"step": 37200
|
2654 |
+
},
|
2655 |
+
{
|
2656 |
+
"epoch": 5.94896331738437,
|
2657 |
+
"grad_norm": 793443.25,
|
2658 |
+
"learning_rate": 4.2530568846358327e-07,
|
2659 |
+
"loss": 3.4279,
|
2660 |
+
"step": 37300
|
2661 |
+
},
|
2662 |
+
{
|
2663 |
+
"epoch": 5.964912280701754,
|
2664 |
+
"grad_norm": 828328.5625,
|
2665 |
+
"learning_rate": 2.9239766081871344e-07,
|
2666 |
+
"loss": 3.4171,
|
2667 |
+
"step": 37400
|
2668 |
+
},
|
2669 |
+
{
|
2670 |
+
"epoch": 5.980861244019139,
|
2671 |
+
"grad_norm": 826741.5625,
|
2672 |
+
"learning_rate": 1.5948963317384372e-07,
|
2673 |
+
"loss": 3.5064,
|
2674 |
+
"step": 37500
|
2675 |
+
},
|
2676 |
+
{
|
2677 |
+
"epoch": 5.996810207336523,
|
2678 |
+
"grad_norm": 698366.375,
|
2679 |
+
"learning_rate": 2.6581605528973954e-08,
|
2680 |
+
"loss": 3.42,
|
2681 |
+
"step": 37600
|
2682 |
+
},
|
2683 |
+
{
|
2684 |
+
"epoch": 6.0,
|
2685 |
+
"eval_loss": 3.615572929382324,
|
2686 |
+
"eval_runtime": 213.1542,
|
2687 |
+
"eval_samples_per_second": 134.283,
|
2688 |
+
"eval_steps_per_second": 4.199,
|
2689 |
+
"step": 37620
|
2690 |
+
}
|
2691 |
+
],
|
2692 |
+
"logging_steps": 100,
|
2693 |
+
"max_steps": 37620,
|
2694 |
+
"num_input_tokens_seen": 0,
|
2695 |
+
"num_train_epochs": 6,
|
2696 |
+
"save_steps": 500,
|
2697 |
+
"stateful_callbacks": {
|
2698 |
+
"TrainerControl": {
|
2699 |
+
"args": {
|
2700 |
+
"should_epoch_stop": false,
|
2701 |
+
"should_evaluate": false,
|
2702 |
+
"should_log": false,
|
2703 |
+
"should_save": true,
|
2704 |
+
"should_training_stop": true
|
2705 |
+
},
|
2706 |
+
"attributes": {}
|
2707 |
+
}
|
2708 |
+
},
|
2709 |
+
"total_flos": 4.65721373251031e+16,
|
2710 |
+
"train_batch_size": 32,
|
2711 |
+
"trial_name": null,
|
2712 |
+
"trial_params": null
|
2713 |
+
}
|
checkpoint-37620/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7e6d8905717aeff0bee4d2b97e0432173098aa3c32a8f4e99a68f0803ee335cd
|
3 |
+
size 5240
|
checkpoint-37620/vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
config.json
ADDED
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "Ransaka/sinhala-bert-medium-v2",
|
3 |
+
"architectures": [
|
4 |
+
"BertForMaskedLM"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_dropout": null,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_dropout_prob": 0.1,
|
10 |
+
"hidden_size": 786,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 1024,
|
13 |
+
"layer_norm_eps": 1e-12,
|
14 |
+
"max_position_embeddings": 256,
|
15 |
+
"model_type": "bert",
|
16 |
+
"num_attention_heads": 6,
|
17 |
+
"num_hidden_layers": 6,
|
18 |
+
"pad_token_id": 0,
|
19 |
+
"position_embedding_type": "absolute",
|
20 |
+
"torch_dtype": "float32",
|
21 |
+
"transformers_version": "4.47.0",
|
22 |
+
"type_vocab_size": 2,
|
23 |
+
"use_cache": true,
|
24 |
+
"vocab_size": 32000
|
25 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"pad_token_id": 0,
|
4 |
+
"transformers_version": "4.47.0"
|
5 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:51874da6e509fec39e26442cbb403925b98087382b3935fe751f465f12a14781
|
3 |
+
size 202182680
|
special_tokens_map.json
ADDED
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": {
|
3 |
+
"content": "[CLS]",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"mask_token": {
|
10 |
+
"content": "[MASK]",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"pad_token": {
|
17 |
+
"content": "[PAD]",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"sep_token": {
|
24 |
+
"content": "[SEP]",
|
25 |
+
"lstrip": false,
|
26 |
+
"normalized": false,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
},
|
30 |
+
"unk_token": {
|
31 |
+
"content": "[UNK]",
|
32 |
+
"lstrip": false,
|
33 |
+
"normalized": false,
|
34 |
+
"rstrip": false,
|
35 |
+
"single_word": false
|
36 |
+
}
|
37 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "[PAD]",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "[UNK]",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "[CLS]",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "[SEP]",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"4": {
|
36 |
+
"content": "[MASK]",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"clean_up_tokenization_spaces": true,
|
45 |
+
"cls_token": "[CLS]",
|
46 |
+
"do_lower_case": false,
|
47 |
+
"extra_special_tokens": {},
|
48 |
+
"mask_token": "[MASK]",
|
49 |
+
"model_max_length": 512,
|
50 |
+
"pad_token": "[PAD]",
|
51 |
+
"sep_token": "[SEP]",
|
52 |
+
"strip_accents": null,
|
53 |
+
"tokenize_chinese_chars": true,
|
54 |
+
"tokenizer_class": "BertTokenizer",
|
55 |
+
"unk_token": "[UNK]"
|
56 |
+
}
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|