Upload folder using huggingface_hub
Browse files- checkpoint-1692/config.json +25 -0
- checkpoint-1692/model.safetensors +3 -0
- checkpoint-1692/optimizer.pt +3 -0
- checkpoint-1692/rng_state.pth +3 -0
- checkpoint-1692/scheduler.pt +3 -0
- checkpoint-1692/special_tokens_map.json +7 -0
- checkpoint-1692/tokenizer.json +0 -0
- checkpoint-1692/tokenizer_config.json +56 -0
- checkpoint-1692/trainer_state.json +1240 -0
- checkpoint-1692/training_args.bin +3 -0
- checkpoint-1692/vocab.txt +0 -0
- config.json +25 -0
- model.safetensors +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +56 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
checkpoint-1692/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "distilbert-base-uncased",
|
| 3 |
+
"activation": "gelu",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"DistilBertForSequenceClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.1,
|
| 8 |
+
"dim": 768,
|
| 9 |
+
"dropout": 0.1,
|
| 10 |
+
"hidden_dim": 3072,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"max_position_embeddings": 512,
|
| 13 |
+
"model_type": "distilbert",
|
| 14 |
+
"n_heads": 12,
|
| 15 |
+
"n_layers": 6,
|
| 16 |
+
"pad_token_id": 0,
|
| 17 |
+
"problem_type": "single_label_classification",
|
| 18 |
+
"qa_dropout": 0.1,
|
| 19 |
+
"seq_classif_dropout": 0.2,
|
| 20 |
+
"sinusoidal_pos_embds": false,
|
| 21 |
+
"tie_weights_": true,
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.48.0",
|
| 24 |
+
"vocab_size": 30522
|
| 25 |
+
}
|
checkpoint-1692/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b232355fadb9bbfb707eed2b6980060355c8d8c7e4b5dc79d821f4dd37af8d23
|
| 3 |
+
size 267832560
|
checkpoint-1692/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ca9edab0ecf2b9d2edb966bbbee452b11e0868b7714b2ef944adbeb30b31472
|
| 3 |
+
size 535727290
|
checkpoint-1692/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7dfc3241a5237ebe6101d857d6ea2aef20fc76c54f21d994c1489138f86373cd
|
| 3 |
+
size 14244
|
checkpoint-1692/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2146e450e271a8b076696e032564edc6a54af58183cc1595f4414249e96aa7c5
|
| 3 |
+
size 1064
|
checkpoint-1692/special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
checkpoint-1692/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1692/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 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 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": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 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": "DistilBertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
checkpoint-1692/trainer_state.json
ADDED
|
@@ -0,0 +1,1240 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1692,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01773049645390071,
|
| 13 |
+
"grad_norm": 1.0585654973983765,
|
| 14 |
+
"learning_rate": 4.970449172576833e-05,
|
| 15 |
+
"loss": 0.4344,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03546099290780142,
|
| 20 |
+
"grad_norm": 0.5365344285964966,
|
| 21 |
+
"learning_rate": 4.940898345153664e-05,
|
| 22 |
+
"loss": 0.357,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05319148936170213,
|
| 27 |
+
"grad_norm": 1.0602149963378906,
|
| 28 |
+
"learning_rate": 4.911347517730497e-05,
|
| 29 |
+
"loss": 0.3397,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.07092198581560284,
|
| 34 |
+
"grad_norm": 2.197633743286133,
|
| 35 |
+
"learning_rate": 4.8817966903073283e-05,
|
| 36 |
+
"loss": 0.3472,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.08865248226950355,
|
| 41 |
+
"grad_norm": 1.7486677169799805,
|
| 42 |
+
"learning_rate": 4.852245862884161e-05,
|
| 43 |
+
"loss": 0.3175,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.10638297872340426,
|
| 48 |
+
"grad_norm": 2.9779016971588135,
|
| 49 |
+
"learning_rate": 4.822695035460993e-05,
|
| 50 |
+
"loss": 0.3174,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.12411347517730496,
|
| 55 |
+
"grad_norm": 1.5208463668823242,
|
| 56 |
+
"learning_rate": 4.793144208037825e-05,
|
| 57 |
+
"loss": 0.3061,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.14184397163120568,
|
| 62 |
+
"grad_norm": 2.6333279609680176,
|
| 63 |
+
"learning_rate": 4.763593380614658e-05,
|
| 64 |
+
"loss": 0.3196,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1595744680851064,
|
| 69 |
+
"grad_norm": 2.9724740982055664,
|
| 70 |
+
"learning_rate": 4.734042553191489e-05,
|
| 71 |
+
"loss": 0.2976,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.1773049645390071,
|
| 76 |
+
"grad_norm": 1.83737313747406,
|
| 77 |
+
"learning_rate": 4.704491725768322e-05,
|
| 78 |
+
"loss": 0.2861,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.1950354609929078,
|
| 83 |
+
"grad_norm": 2.455568313598633,
|
| 84 |
+
"learning_rate": 4.674940898345154e-05,
|
| 85 |
+
"loss": 0.3028,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2127659574468085,
|
| 90 |
+
"grad_norm": 1.068047046661377,
|
| 91 |
+
"learning_rate": 4.645390070921986e-05,
|
| 92 |
+
"loss": 0.29,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.23049645390070922,
|
| 97 |
+
"grad_norm": 1.9365094900131226,
|
| 98 |
+
"learning_rate": 4.615839243498818e-05,
|
| 99 |
+
"loss": 0.2338,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.24822695035460993,
|
| 104 |
+
"grad_norm": 4.382534027099609,
|
| 105 |
+
"learning_rate": 4.58628841607565e-05,
|
| 106 |
+
"loss": 0.2383,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.26595744680851063,
|
| 111 |
+
"grad_norm": 1.8602826595306396,
|
| 112 |
+
"learning_rate": 4.556737588652483e-05,
|
| 113 |
+
"loss": 0.2698,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.28368794326241137,
|
| 118 |
+
"grad_norm": 4.105195999145508,
|
| 119 |
+
"learning_rate": 4.527186761229315e-05,
|
| 120 |
+
"loss": 0.2895,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.30141843971631205,
|
| 125 |
+
"grad_norm": 2.267707347869873,
|
| 126 |
+
"learning_rate": 4.497635933806147e-05,
|
| 127 |
+
"loss": 0.3412,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.3191489361702128,
|
| 132 |
+
"grad_norm": 1.7341336011886597,
|
| 133 |
+
"learning_rate": 4.468085106382979e-05,
|
| 134 |
+
"loss": 0.2304,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.33687943262411346,
|
| 139 |
+
"grad_norm": 4.169299602508545,
|
| 140 |
+
"learning_rate": 4.438534278959811e-05,
|
| 141 |
+
"loss": 0.2063,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.3546099290780142,
|
| 146 |
+
"grad_norm": 4.8010663986206055,
|
| 147 |
+
"learning_rate": 4.4089834515366435e-05,
|
| 148 |
+
"loss": 0.2919,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.3723404255319149,
|
| 153 |
+
"grad_norm": 1.2835743427276611,
|
| 154 |
+
"learning_rate": 4.3794326241134755e-05,
|
| 155 |
+
"loss": 0.2172,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.3900709219858156,
|
| 160 |
+
"grad_norm": 2.9821650981903076,
|
| 161 |
+
"learning_rate": 4.3498817966903076e-05,
|
| 162 |
+
"loss": 0.2404,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.4078014184397163,
|
| 167 |
+
"grad_norm": 4.525284767150879,
|
| 168 |
+
"learning_rate": 4.3203309692671396e-05,
|
| 169 |
+
"loss": 0.2233,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.425531914893617,
|
| 174 |
+
"grad_norm": 2.9262726306915283,
|
| 175 |
+
"learning_rate": 4.2907801418439716e-05,
|
| 176 |
+
"loss": 0.2247,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.4432624113475177,
|
| 181 |
+
"grad_norm": 1.120538592338562,
|
| 182 |
+
"learning_rate": 4.2612293144208036e-05,
|
| 183 |
+
"loss": 0.2,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.46099290780141844,
|
| 188 |
+
"grad_norm": 3.67739200592041,
|
| 189 |
+
"learning_rate": 4.231678486997636e-05,
|
| 190 |
+
"loss": 0.2774,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.4787234042553192,
|
| 195 |
+
"grad_norm": 2.233776330947876,
|
| 196 |
+
"learning_rate": 4.2021276595744684e-05,
|
| 197 |
+
"loss": 0.2339,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.49645390070921985,
|
| 202 |
+
"grad_norm": 3.464088201522827,
|
| 203 |
+
"learning_rate": 4.1725768321513004e-05,
|
| 204 |
+
"loss": 0.2443,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.5141843971631206,
|
| 209 |
+
"grad_norm": 3.5438666343688965,
|
| 210 |
+
"learning_rate": 4.1430260047281324e-05,
|
| 211 |
+
"loss": 0.2949,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.5319148936170213,
|
| 216 |
+
"grad_norm": 2.237935781478882,
|
| 217 |
+
"learning_rate": 4.1134751773049644e-05,
|
| 218 |
+
"loss": 0.212,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.549645390070922,
|
| 223 |
+
"grad_norm": 2.065207004547119,
|
| 224 |
+
"learning_rate": 4.083924349881797e-05,
|
| 225 |
+
"loss": 0.1957,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.5673758865248227,
|
| 230 |
+
"grad_norm": 1.892553448677063,
|
| 231 |
+
"learning_rate": 4.0543735224586285e-05,
|
| 232 |
+
"loss": 0.2182,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.5851063829787234,
|
| 237 |
+
"grad_norm": 1.1844576597213745,
|
| 238 |
+
"learning_rate": 4.024822695035461e-05,
|
| 239 |
+
"loss": 0.258,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.6028368794326241,
|
| 244 |
+
"grad_norm": 1.786897897720337,
|
| 245 |
+
"learning_rate": 3.995271867612293e-05,
|
| 246 |
+
"loss": 0.2151,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.6205673758865248,
|
| 251 |
+
"grad_norm": 4.924149036407471,
|
| 252 |
+
"learning_rate": 3.965721040189125e-05,
|
| 253 |
+
"loss": 0.2095,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.6382978723404256,
|
| 258 |
+
"grad_norm": 3.0387415885925293,
|
| 259 |
+
"learning_rate": 3.936170212765958e-05,
|
| 260 |
+
"loss": 0.2339,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.6560283687943262,
|
| 265 |
+
"grad_norm": 0.9237979054450989,
|
| 266 |
+
"learning_rate": 3.906619385342789e-05,
|
| 267 |
+
"loss": 0.2731,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.6737588652482269,
|
| 272 |
+
"grad_norm": 1.2017602920532227,
|
| 273 |
+
"learning_rate": 3.877068557919622e-05,
|
| 274 |
+
"loss": 0.2323,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.6914893617021277,
|
| 279 |
+
"grad_norm": 4.876067161560059,
|
| 280 |
+
"learning_rate": 3.847517730496454e-05,
|
| 281 |
+
"loss": 0.2312,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.7092198581560284,
|
| 286 |
+
"grad_norm": 2.4507224559783936,
|
| 287 |
+
"learning_rate": 3.817966903073286e-05,
|
| 288 |
+
"loss": 0.2548,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.7269503546099291,
|
| 293 |
+
"grad_norm": 3.6143534183502197,
|
| 294 |
+
"learning_rate": 3.788416075650119e-05,
|
| 295 |
+
"loss": 0.2564,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.7446808510638298,
|
| 300 |
+
"grad_norm": 3.5805842876434326,
|
| 301 |
+
"learning_rate": 3.75886524822695e-05,
|
| 302 |
+
"loss": 0.2275,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.7624113475177305,
|
| 307 |
+
"grad_norm": 3.5016636848449707,
|
| 308 |
+
"learning_rate": 3.729314420803783e-05,
|
| 309 |
+
"loss": 0.2724,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.7801418439716312,
|
| 314 |
+
"grad_norm": 4.354279518127441,
|
| 315 |
+
"learning_rate": 3.699763593380615e-05,
|
| 316 |
+
"loss": 0.1945,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.7978723404255319,
|
| 321 |
+
"grad_norm": 1.473893642425537,
|
| 322 |
+
"learning_rate": 3.670212765957447e-05,
|
| 323 |
+
"loss": 0.2098,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.8156028368794326,
|
| 328 |
+
"grad_norm": 3.5749127864837646,
|
| 329 |
+
"learning_rate": 3.6406619385342796e-05,
|
| 330 |
+
"loss": 0.2062,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.8333333333333334,
|
| 335 |
+
"grad_norm": 1.5818848609924316,
|
| 336 |
+
"learning_rate": 3.611111111111111e-05,
|
| 337 |
+
"loss": 0.226,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.851063829787234,
|
| 342 |
+
"grad_norm": 0.9678372144699097,
|
| 343 |
+
"learning_rate": 3.5815602836879437e-05,
|
| 344 |
+
"loss": 0.2087,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.8687943262411347,
|
| 349 |
+
"grad_norm": 3.465823173522949,
|
| 350 |
+
"learning_rate": 3.552009456264776e-05,
|
| 351 |
+
"loss": 0.2473,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.8865248226950354,
|
| 356 |
+
"grad_norm": 2.4084010124206543,
|
| 357 |
+
"learning_rate": 3.522458628841608e-05,
|
| 358 |
+
"loss": 0.2732,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.9042553191489362,
|
| 363 |
+
"grad_norm": 2.5898969173431396,
|
| 364 |
+
"learning_rate": 3.49290780141844e-05,
|
| 365 |
+
"loss": 0.1784,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.9219858156028369,
|
| 370 |
+
"grad_norm": 2.0226316452026367,
|
| 371 |
+
"learning_rate": 3.463356973995272e-05,
|
| 372 |
+
"loss": 0.1835,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.9397163120567376,
|
| 377 |
+
"grad_norm": 2.84747576713562,
|
| 378 |
+
"learning_rate": 3.4338061465721045e-05,
|
| 379 |
+
"loss": 0.2101,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.9574468085106383,
|
| 384 |
+
"grad_norm": 1.7862353324890137,
|
| 385 |
+
"learning_rate": 3.4042553191489365e-05,
|
| 386 |
+
"loss": 0.1742,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.975177304964539,
|
| 391 |
+
"grad_norm": 5.6382155418396,
|
| 392 |
+
"learning_rate": 3.3747044917257685e-05,
|
| 393 |
+
"loss": 0.1596,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.9929078014184397,
|
| 398 |
+
"grad_norm": 2.4179556369781494,
|
| 399 |
+
"learning_rate": 3.3451536643026005e-05,
|
| 400 |
+
"loss": 0.2326,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.0,
|
| 405 |
+
"eval_accuracy": 0.9245283018867925,
|
| 406 |
+
"eval_f1": 0.4970414201183432,
|
| 407 |
+
"eval_loss": 0.1933222860097885,
|
| 408 |
+
"eval_precision": 0.7924528301886793,
|
| 409 |
+
"eval_recall": 0.3620689655172414,
|
| 410 |
+
"eval_runtime": 18.2335,
|
| 411 |
+
"eval_samples_per_second": 247.073,
|
| 412 |
+
"eval_steps_per_second": 30.932,
|
| 413 |
+
"step": 564
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.0106382978723405,
|
| 417 |
+
"grad_norm": 1.8270294666290283,
|
| 418 |
+
"learning_rate": 3.3156028368794326e-05,
|
| 419 |
+
"loss": 0.1533,
|
| 420 |
+
"step": 570
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 1.0283687943262412,
|
| 424 |
+
"grad_norm": 2.191605567932129,
|
| 425 |
+
"learning_rate": 3.2860520094562646e-05,
|
| 426 |
+
"loss": 0.1668,
|
| 427 |
+
"step": 580
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.0460992907801419,
|
| 431 |
+
"grad_norm": 5.722381591796875,
|
| 432 |
+
"learning_rate": 3.256501182033097e-05,
|
| 433 |
+
"loss": 0.1999,
|
| 434 |
+
"step": 590
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.0638297872340425,
|
| 438 |
+
"grad_norm": 5.328127384185791,
|
| 439 |
+
"learning_rate": 3.226950354609929e-05,
|
| 440 |
+
"loss": 0.1799,
|
| 441 |
+
"step": 600
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.0815602836879432,
|
| 445 |
+
"grad_norm": 3.9988811016082764,
|
| 446 |
+
"learning_rate": 3.1973995271867614e-05,
|
| 447 |
+
"loss": 0.1787,
|
| 448 |
+
"step": 610
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 1.099290780141844,
|
| 452 |
+
"grad_norm": 2.0374648571014404,
|
| 453 |
+
"learning_rate": 3.1678486997635934e-05,
|
| 454 |
+
"loss": 0.1728,
|
| 455 |
+
"step": 620
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.1170212765957448,
|
| 459 |
+
"grad_norm": 1.21236252784729,
|
| 460 |
+
"learning_rate": 3.1382978723404254e-05,
|
| 461 |
+
"loss": 0.1884,
|
| 462 |
+
"step": 630
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 1.1347517730496455,
|
| 466 |
+
"grad_norm": 1.166555404663086,
|
| 467 |
+
"learning_rate": 3.108747044917258e-05,
|
| 468 |
+
"loss": 0.1946,
|
| 469 |
+
"step": 640
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 1.1524822695035462,
|
| 473 |
+
"grad_norm": 1.6630871295928955,
|
| 474 |
+
"learning_rate": 3.0791962174940895e-05,
|
| 475 |
+
"loss": 0.1791,
|
| 476 |
+
"step": 650
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.1702127659574468,
|
| 480 |
+
"grad_norm": 1.3428844213485718,
|
| 481 |
+
"learning_rate": 3.0496453900709222e-05,
|
| 482 |
+
"loss": 0.1607,
|
| 483 |
+
"step": 660
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.1879432624113475,
|
| 487 |
+
"grad_norm": 2.1707041263580322,
|
| 488 |
+
"learning_rate": 3.0200945626477545e-05,
|
| 489 |
+
"loss": 0.1888,
|
| 490 |
+
"step": 670
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 1.2056737588652482,
|
| 494 |
+
"grad_norm": 2.2633979320526123,
|
| 495 |
+
"learning_rate": 2.9905437352245862e-05,
|
| 496 |
+
"loss": 0.159,
|
| 497 |
+
"step": 680
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.2234042553191489,
|
| 501 |
+
"grad_norm": 2.3307528495788574,
|
| 502 |
+
"learning_rate": 2.9609929078014186e-05,
|
| 503 |
+
"loss": 0.145,
|
| 504 |
+
"step": 690
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.2411347517730495,
|
| 508 |
+
"grad_norm": 5.760026931762695,
|
| 509 |
+
"learning_rate": 2.9314420803782506e-05,
|
| 510 |
+
"loss": 0.1352,
|
| 511 |
+
"step": 700
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.2588652482269502,
|
| 515 |
+
"grad_norm": 2.038200855255127,
|
| 516 |
+
"learning_rate": 2.901891252955083e-05,
|
| 517 |
+
"loss": 0.1637,
|
| 518 |
+
"step": 710
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.2765957446808511,
|
| 522 |
+
"grad_norm": 3.5735347270965576,
|
| 523 |
+
"learning_rate": 2.8723404255319154e-05,
|
| 524 |
+
"loss": 0.125,
|
| 525 |
+
"step": 720
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.2943262411347518,
|
| 529 |
+
"grad_norm": 1.2548505067825317,
|
| 530 |
+
"learning_rate": 2.842789598108747e-05,
|
| 531 |
+
"loss": 0.1571,
|
| 532 |
+
"step": 730
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.3120567375886525,
|
| 536 |
+
"grad_norm": 3.3048200607299805,
|
| 537 |
+
"learning_rate": 2.8132387706855794e-05,
|
| 538 |
+
"loss": 0.1298,
|
| 539 |
+
"step": 740
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 1.3297872340425532,
|
| 543 |
+
"grad_norm": 1.5077462196350098,
|
| 544 |
+
"learning_rate": 2.7836879432624114e-05,
|
| 545 |
+
"loss": 0.1593,
|
| 546 |
+
"step": 750
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 1.3475177304964538,
|
| 550 |
+
"grad_norm": 2.6117773056030273,
|
| 551 |
+
"learning_rate": 2.7541371158392438e-05,
|
| 552 |
+
"loss": 0.2071,
|
| 553 |
+
"step": 760
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.3652482269503547,
|
| 557 |
+
"grad_norm": 3.2539641857147217,
|
| 558 |
+
"learning_rate": 2.7245862884160755e-05,
|
| 559 |
+
"loss": 0.1661,
|
| 560 |
+
"step": 770
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.3829787234042552,
|
| 564 |
+
"grad_norm": 2.994198799133301,
|
| 565 |
+
"learning_rate": 2.695035460992908e-05,
|
| 566 |
+
"loss": 0.1904,
|
| 567 |
+
"step": 780
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 1.400709219858156,
|
| 571 |
+
"grad_norm": 2.030109405517578,
|
| 572 |
+
"learning_rate": 2.6654846335697402e-05,
|
| 573 |
+
"loss": 0.1583,
|
| 574 |
+
"step": 790
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 1.4184397163120568,
|
| 578 |
+
"grad_norm": 1.1124039888381958,
|
| 579 |
+
"learning_rate": 2.6359338061465723e-05,
|
| 580 |
+
"loss": 0.2241,
|
| 581 |
+
"step": 800
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 1.4361702127659575,
|
| 585 |
+
"grad_norm": 1.3373945951461792,
|
| 586 |
+
"learning_rate": 2.6063829787234046e-05,
|
| 587 |
+
"loss": 0.1659,
|
| 588 |
+
"step": 810
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 1.4539007092198581,
|
| 592 |
+
"grad_norm": 4.298996448516846,
|
| 593 |
+
"learning_rate": 2.5768321513002363e-05,
|
| 594 |
+
"loss": 0.1597,
|
| 595 |
+
"step": 820
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 1.4716312056737588,
|
| 599 |
+
"grad_norm": 2.2879459857940674,
|
| 600 |
+
"learning_rate": 2.5472813238770687e-05,
|
| 601 |
+
"loss": 0.1806,
|
| 602 |
+
"step": 830
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.4893617021276595,
|
| 606 |
+
"grad_norm": 5.475185394287109,
|
| 607 |
+
"learning_rate": 2.5177304964539007e-05,
|
| 608 |
+
"loss": 0.2264,
|
| 609 |
+
"step": 840
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.5070921985815602,
|
| 613 |
+
"grad_norm": 3.100987672805786,
|
| 614 |
+
"learning_rate": 2.488179669030733e-05,
|
| 615 |
+
"loss": 0.1517,
|
| 616 |
+
"step": 850
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 1.524822695035461,
|
| 620 |
+
"grad_norm": 1.2688792943954468,
|
| 621 |
+
"learning_rate": 2.458628841607565e-05,
|
| 622 |
+
"loss": 0.1257,
|
| 623 |
+
"step": 860
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 1.5425531914893615,
|
| 627 |
+
"grad_norm": 2.6867027282714844,
|
| 628 |
+
"learning_rate": 2.429078014184397e-05,
|
| 629 |
+
"loss": 0.1489,
|
| 630 |
+
"step": 870
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 1.5602836879432624,
|
| 634 |
+
"grad_norm": 2.5701425075531006,
|
| 635 |
+
"learning_rate": 2.3995271867612295e-05,
|
| 636 |
+
"loss": 0.2134,
|
| 637 |
+
"step": 880
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 1.5780141843971631,
|
| 641 |
+
"grad_norm": 5.73723840713501,
|
| 642 |
+
"learning_rate": 2.3699763593380615e-05,
|
| 643 |
+
"loss": 0.191,
|
| 644 |
+
"step": 890
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 1.5957446808510638,
|
| 648 |
+
"grad_norm": 3.5702483654022217,
|
| 649 |
+
"learning_rate": 2.340425531914894e-05,
|
| 650 |
+
"loss": 0.17,
|
| 651 |
+
"step": 900
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 1.6134751773049647,
|
| 655 |
+
"grad_norm": 4.70973014831543,
|
| 656 |
+
"learning_rate": 2.310874704491726e-05,
|
| 657 |
+
"loss": 0.1357,
|
| 658 |
+
"step": 910
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 1.6312056737588652,
|
| 662 |
+
"grad_norm": 1.4206900596618652,
|
| 663 |
+
"learning_rate": 2.281323877068558e-05,
|
| 664 |
+
"loss": 0.1264,
|
| 665 |
+
"step": 920
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 1.648936170212766,
|
| 669 |
+
"grad_norm": 3.2121095657348633,
|
| 670 |
+
"learning_rate": 2.25177304964539e-05,
|
| 671 |
+
"loss": 0.1409,
|
| 672 |
+
"step": 930
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 1.6666666666666665,
|
| 676 |
+
"grad_norm": 5.003275394439697,
|
| 677 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 678 |
+
"loss": 0.1993,
|
| 679 |
+
"step": 940
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 1.6843971631205674,
|
| 683 |
+
"grad_norm": 8.793657302856445,
|
| 684 |
+
"learning_rate": 2.1926713947990547e-05,
|
| 685 |
+
"loss": 0.097,
|
| 686 |
+
"step": 950
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 1.702127659574468,
|
| 690 |
+
"grad_norm": 2.719095230102539,
|
| 691 |
+
"learning_rate": 2.1631205673758867e-05,
|
| 692 |
+
"loss": 0.1739,
|
| 693 |
+
"step": 960
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 1.7198581560283688,
|
| 697 |
+
"grad_norm": 4.192145347595215,
|
| 698 |
+
"learning_rate": 2.1335697399527187e-05,
|
| 699 |
+
"loss": 0.1819,
|
| 700 |
+
"step": 970
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 1.7375886524822695,
|
| 704 |
+
"grad_norm": 3.239504098892212,
|
| 705 |
+
"learning_rate": 2.1040189125295508e-05,
|
| 706 |
+
"loss": 0.1775,
|
| 707 |
+
"step": 980
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 1.7553191489361701,
|
| 711 |
+
"grad_norm": 1.6205350160598755,
|
| 712 |
+
"learning_rate": 2.074468085106383e-05,
|
| 713 |
+
"loss": 0.1455,
|
| 714 |
+
"step": 990
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 1.773049645390071,
|
| 718 |
+
"grad_norm": 2.7471115589141846,
|
| 719 |
+
"learning_rate": 2.0449172576832152e-05,
|
| 720 |
+
"loss": 0.2198,
|
| 721 |
+
"step": 1000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 1.7907801418439715,
|
| 725 |
+
"grad_norm": 2.1342594623565674,
|
| 726 |
+
"learning_rate": 2.0153664302600475e-05,
|
| 727 |
+
"loss": 0.1577,
|
| 728 |
+
"step": 1010
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 1.8085106382978724,
|
| 732 |
+
"grad_norm": 2.6111040115356445,
|
| 733 |
+
"learning_rate": 1.9858156028368796e-05,
|
| 734 |
+
"loss": 0.1907,
|
| 735 |
+
"step": 1020
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 1.826241134751773,
|
| 739 |
+
"grad_norm": 1.9335544109344482,
|
| 740 |
+
"learning_rate": 1.9562647754137116e-05,
|
| 741 |
+
"loss": 0.1812,
|
| 742 |
+
"step": 1030
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 1.8439716312056738,
|
| 746 |
+
"grad_norm": 3.2039554119110107,
|
| 747 |
+
"learning_rate": 1.926713947990544e-05,
|
| 748 |
+
"loss": 0.1604,
|
| 749 |
+
"step": 1040
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 1.8617021276595744,
|
| 753 |
+
"grad_norm": 4.588708400726318,
|
| 754 |
+
"learning_rate": 1.897163120567376e-05,
|
| 755 |
+
"loss": 0.1576,
|
| 756 |
+
"step": 1050
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 1.8794326241134751,
|
| 760 |
+
"grad_norm": 2.468317985534668,
|
| 761 |
+
"learning_rate": 1.867612293144208e-05,
|
| 762 |
+
"loss": 0.2102,
|
| 763 |
+
"step": 1060
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 1.897163120567376,
|
| 767 |
+
"grad_norm": 2.883124828338623,
|
| 768 |
+
"learning_rate": 1.83806146572104e-05,
|
| 769 |
+
"loss": 0.1883,
|
| 770 |
+
"step": 1070
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 1.9148936170212765,
|
| 774 |
+
"grad_norm": 1.7469326257705688,
|
| 775 |
+
"learning_rate": 1.8085106382978724e-05,
|
| 776 |
+
"loss": 0.1598,
|
| 777 |
+
"step": 1080
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 1.9326241134751774,
|
| 781 |
+
"grad_norm": 3.9554150104522705,
|
| 782 |
+
"learning_rate": 1.7789598108747048e-05,
|
| 783 |
+
"loss": 0.171,
|
| 784 |
+
"step": 1090
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 1.950354609929078,
|
| 788 |
+
"grad_norm": 7.007796287536621,
|
| 789 |
+
"learning_rate": 1.7494089834515368e-05,
|
| 790 |
+
"loss": 0.1908,
|
| 791 |
+
"step": 1100
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 1.9680851063829787,
|
| 795 |
+
"grad_norm": 2.2939341068267822,
|
| 796 |
+
"learning_rate": 1.7198581560283688e-05,
|
| 797 |
+
"loss": 0.2282,
|
| 798 |
+
"step": 1110
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 1.9858156028368794,
|
| 802 |
+
"grad_norm": 1.1014021635055542,
|
| 803 |
+
"learning_rate": 1.690307328605201e-05,
|
| 804 |
+
"loss": 0.1599,
|
| 805 |
+
"step": 1120
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 2.0,
|
| 809 |
+
"eval_accuracy": 0.93007769145394,
|
| 810 |
+
"eval_f1": 0.5827814569536424,
|
| 811 |
+
"eval_loss": 0.18422859907150269,
|
| 812 |
+
"eval_precision": 0.7560137457044673,
|
| 813 |
+
"eval_recall": 0.47413793103448276,
|
| 814 |
+
"eval_runtime": 18.3061,
|
| 815 |
+
"eval_samples_per_second": 246.093,
|
| 816 |
+
"eval_steps_per_second": 30.809,
|
| 817 |
+
"step": 1128
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 2.00354609929078,
|
| 821 |
+
"grad_norm": 1.147599458694458,
|
| 822 |
+
"learning_rate": 1.6607565011820332e-05,
|
| 823 |
+
"loss": 0.1652,
|
| 824 |
+
"step": 1130
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 2.021276595744681,
|
| 828 |
+
"grad_norm": 1.893280029296875,
|
| 829 |
+
"learning_rate": 1.6312056737588656e-05,
|
| 830 |
+
"loss": 0.1345,
|
| 831 |
+
"step": 1140
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 2.0390070921985815,
|
| 835 |
+
"grad_norm": 3.6475846767425537,
|
| 836 |
+
"learning_rate": 1.6016548463356976e-05,
|
| 837 |
+
"loss": 0.1214,
|
| 838 |
+
"step": 1150
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 2.0567375886524824,
|
| 842 |
+
"grad_norm": 6.175610542297363,
|
| 843 |
+
"learning_rate": 1.5721040189125296e-05,
|
| 844 |
+
"loss": 0.131,
|
| 845 |
+
"step": 1160
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 2.074468085106383,
|
| 849 |
+
"grad_norm": 2.870246171951294,
|
| 850 |
+
"learning_rate": 1.5425531914893617e-05,
|
| 851 |
+
"loss": 0.1379,
|
| 852 |
+
"step": 1170
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 2.0921985815602837,
|
| 856 |
+
"grad_norm": 1.1721495389938354,
|
| 857 |
+
"learning_rate": 1.5130023640661939e-05,
|
| 858 |
+
"loss": 0.1239,
|
| 859 |
+
"step": 1180
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 2.1099290780141846,
|
| 863 |
+
"grad_norm": 5.15436315536499,
|
| 864 |
+
"learning_rate": 1.483451536643026e-05,
|
| 865 |
+
"loss": 0.097,
|
| 866 |
+
"step": 1190
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 2.127659574468085,
|
| 870 |
+
"grad_norm": 0.1995478719472885,
|
| 871 |
+
"learning_rate": 1.4539007092198581e-05,
|
| 872 |
+
"loss": 0.1416,
|
| 873 |
+
"step": 1200
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 2.145390070921986,
|
| 877 |
+
"grad_norm": 4.829070091247559,
|
| 878 |
+
"learning_rate": 1.4243498817966905e-05,
|
| 879 |
+
"loss": 0.1075,
|
| 880 |
+
"step": 1210
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 2.1631205673758864,
|
| 884 |
+
"grad_norm": 4.2572479248046875,
|
| 885 |
+
"learning_rate": 1.3947990543735227e-05,
|
| 886 |
+
"loss": 0.1233,
|
| 887 |
+
"step": 1220
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 2.1808510638297873,
|
| 891 |
+
"grad_norm": 3.297318458557129,
|
| 892 |
+
"learning_rate": 1.3652482269503547e-05,
|
| 893 |
+
"loss": 0.1171,
|
| 894 |
+
"step": 1230
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 2.198581560283688,
|
| 898 |
+
"grad_norm": 2.2237844467163086,
|
| 899 |
+
"learning_rate": 1.3356973995271869e-05,
|
| 900 |
+
"loss": 0.0868,
|
| 901 |
+
"step": 1240
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 2.2163120567375887,
|
| 905 |
+
"grad_norm": 4.605975151062012,
|
| 906 |
+
"learning_rate": 1.3061465721040189e-05,
|
| 907 |
+
"loss": 0.1458,
|
| 908 |
+
"step": 1250
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 2.2340425531914896,
|
| 912 |
+
"grad_norm": 6.8933610916137695,
|
| 913 |
+
"learning_rate": 1.2765957446808511e-05,
|
| 914 |
+
"loss": 0.1242,
|
| 915 |
+
"step": 1260
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 2.25177304964539,
|
| 919 |
+
"grad_norm": 2.5259997844696045,
|
| 920 |
+
"learning_rate": 1.2470449172576833e-05,
|
| 921 |
+
"loss": 0.1131,
|
| 922 |
+
"step": 1270
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 2.269503546099291,
|
| 926 |
+
"grad_norm": 5.3347296714782715,
|
| 927 |
+
"learning_rate": 1.2174940898345153e-05,
|
| 928 |
+
"loss": 0.1334,
|
| 929 |
+
"step": 1280
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 2.2872340425531914,
|
| 933 |
+
"grad_norm": 3.3900346755981445,
|
| 934 |
+
"learning_rate": 1.1879432624113477e-05,
|
| 935 |
+
"loss": 0.1162,
|
| 936 |
+
"step": 1290
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 2.3049645390070923,
|
| 940 |
+
"grad_norm": 3.4547274112701416,
|
| 941 |
+
"learning_rate": 1.1583924349881797e-05,
|
| 942 |
+
"loss": 0.1586,
|
| 943 |
+
"step": 1300
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 2.3226950354609928,
|
| 947 |
+
"grad_norm": 3.0310921669006348,
|
| 948 |
+
"learning_rate": 1.1288416075650119e-05,
|
| 949 |
+
"loss": 0.1088,
|
| 950 |
+
"step": 1310
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 2.3404255319148937,
|
| 954 |
+
"grad_norm": 4.557296276092529,
|
| 955 |
+
"learning_rate": 1.0992907801418441e-05,
|
| 956 |
+
"loss": 0.0988,
|
| 957 |
+
"step": 1320
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 2.3581560283687946,
|
| 961 |
+
"grad_norm": 18.163665771484375,
|
| 962 |
+
"learning_rate": 1.0697399527186761e-05,
|
| 963 |
+
"loss": 0.1055,
|
| 964 |
+
"step": 1330
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 2.375886524822695,
|
| 968 |
+
"grad_norm": 3.0445291996002197,
|
| 969 |
+
"learning_rate": 1.0401891252955083e-05,
|
| 970 |
+
"loss": 0.1547,
|
| 971 |
+
"step": 1340
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 2.393617021276596,
|
| 975 |
+
"grad_norm": 2.7467446327209473,
|
| 976 |
+
"learning_rate": 1.0106382978723404e-05,
|
| 977 |
+
"loss": 0.1248,
|
| 978 |
+
"step": 1350
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 2.4113475177304964,
|
| 982 |
+
"grad_norm": 0.8503823280334473,
|
| 983 |
+
"learning_rate": 9.810874704491727e-06,
|
| 984 |
+
"loss": 0.0915,
|
| 985 |
+
"step": 1360
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 2.4290780141843973,
|
| 989 |
+
"grad_norm": 1.9473341703414917,
|
| 990 |
+
"learning_rate": 9.515366430260048e-06,
|
| 991 |
+
"loss": 0.1267,
|
| 992 |
+
"step": 1370
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.4468085106382977,
|
| 996 |
+
"grad_norm": 2.457197427749634,
|
| 997 |
+
"learning_rate": 9.219858156028368e-06,
|
| 998 |
+
"loss": 0.1228,
|
| 999 |
+
"step": 1380
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.4645390070921986,
|
| 1003 |
+
"grad_norm": 2.5377461910247803,
|
| 1004 |
+
"learning_rate": 8.924349881796691e-06,
|
| 1005 |
+
"loss": 0.085,
|
| 1006 |
+
"step": 1390
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 2.482269503546099,
|
| 1010 |
+
"grad_norm": 4.247258186340332,
|
| 1011 |
+
"learning_rate": 8.628841607565012e-06,
|
| 1012 |
+
"loss": 0.1345,
|
| 1013 |
+
"step": 1400
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 2.5,
|
| 1017 |
+
"grad_norm": 4.234439373016357,
|
| 1018 |
+
"learning_rate": 8.333333333333334e-06,
|
| 1019 |
+
"loss": 0.1396,
|
| 1020 |
+
"step": 1410
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 2.5177304964539005,
|
| 1024 |
+
"grad_norm": 2.6999399662017822,
|
| 1025 |
+
"learning_rate": 8.037825059101656e-06,
|
| 1026 |
+
"loss": 0.0934,
|
| 1027 |
+
"step": 1420
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 2.5354609929078014,
|
| 1031 |
+
"grad_norm": 3.0950117111206055,
|
| 1032 |
+
"learning_rate": 7.742316784869976e-06,
|
| 1033 |
+
"loss": 0.088,
|
| 1034 |
+
"step": 1430
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 2.5531914893617023,
|
| 1038 |
+
"grad_norm": 6.523775577545166,
|
| 1039 |
+
"learning_rate": 7.446808510638298e-06,
|
| 1040 |
+
"loss": 0.1234,
|
| 1041 |
+
"step": 1440
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 2.5709219858156027,
|
| 1045 |
+
"grad_norm": 0.6427033543586731,
|
| 1046 |
+
"learning_rate": 7.151300236406621e-06,
|
| 1047 |
+
"loss": 0.0849,
|
| 1048 |
+
"step": 1450
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 2.5886524822695036,
|
| 1052 |
+
"grad_norm": 4.271224498748779,
|
| 1053 |
+
"learning_rate": 6.855791962174941e-06,
|
| 1054 |
+
"loss": 0.158,
|
| 1055 |
+
"step": 1460
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 2.6063829787234045,
|
| 1059 |
+
"grad_norm": 1.9938910007476807,
|
| 1060 |
+
"learning_rate": 6.560283687943262e-06,
|
| 1061 |
+
"loss": 0.1214,
|
| 1062 |
+
"step": 1470
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 2.624113475177305,
|
| 1066 |
+
"grad_norm": 1.0301127433776855,
|
| 1067 |
+
"learning_rate": 6.264775413711583e-06,
|
| 1068 |
+
"loss": 0.1217,
|
| 1069 |
+
"step": 1480
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 2.6418439716312054,
|
| 1073 |
+
"grad_norm": 1.3055609464645386,
|
| 1074 |
+
"learning_rate": 5.969267139479906e-06,
|
| 1075 |
+
"loss": 0.0761,
|
| 1076 |
+
"step": 1490
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 2.6595744680851063,
|
| 1080 |
+
"grad_norm": 4.245100021362305,
|
| 1081 |
+
"learning_rate": 5.673758865248227e-06,
|
| 1082 |
+
"loss": 0.0884,
|
| 1083 |
+
"step": 1500
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 2.6773049645390072,
|
| 1087 |
+
"grad_norm": 0.3163594901561737,
|
| 1088 |
+
"learning_rate": 5.378250591016549e-06,
|
| 1089 |
+
"loss": 0.081,
|
| 1090 |
+
"step": 1510
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 2.6950354609929077,
|
| 1094 |
+
"grad_norm": 2.392202615737915,
|
| 1095 |
+
"learning_rate": 5.08274231678487e-06,
|
| 1096 |
+
"loss": 0.1288,
|
| 1097 |
+
"step": 1520
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 2.7127659574468086,
|
| 1101 |
+
"grad_norm": 4.6757493019104,
|
| 1102 |
+
"learning_rate": 4.787234042553191e-06,
|
| 1103 |
+
"loss": 0.1134,
|
| 1104 |
+
"step": 1530
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 2.7304964539007095,
|
| 1108 |
+
"grad_norm": 14.765192985534668,
|
| 1109 |
+
"learning_rate": 4.491725768321513e-06,
|
| 1110 |
+
"loss": 0.1706,
|
| 1111 |
+
"step": 1540
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 2.74822695035461,
|
| 1115 |
+
"grad_norm": 5.195065021514893,
|
| 1116 |
+
"learning_rate": 4.1962174940898345e-06,
|
| 1117 |
+
"loss": 0.1669,
|
| 1118 |
+
"step": 1550
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 2.7659574468085104,
|
| 1122 |
+
"grad_norm": 2.5050220489501953,
|
| 1123 |
+
"learning_rate": 3.9007092198581565e-06,
|
| 1124 |
+
"loss": 0.0757,
|
| 1125 |
+
"step": 1560
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 2.7836879432624113,
|
| 1129 |
+
"grad_norm": 2.086787223815918,
|
| 1130 |
+
"learning_rate": 3.605200945626478e-06,
|
| 1131 |
+
"loss": 0.1797,
|
| 1132 |
+
"step": 1570
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 2.801418439716312,
|
| 1136 |
+
"grad_norm": 5.924837589263916,
|
| 1137 |
+
"learning_rate": 3.309692671394799e-06,
|
| 1138 |
+
"loss": 0.1602,
|
| 1139 |
+
"step": 1580
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 2.8191489361702127,
|
| 1143 |
+
"grad_norm": 4.369917392730713,
|
| 1144 |
+
"learning_rate": 3.0141843971631207e-06,
|
| 1145 |
+
"loss": 0.08,
|
| 1146 |
+
"step": 1590
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 2.8368794326241136,
|
| 1150 |
+
"grad_norm": 1.1417466402053833,
|
| 1151 |
+
"learning_rate": 2.7186761229314422e-06,
|
| 1152 |
+
"loss": 0.1123,
|
| 1153 |
+
"step": 1600
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 2.854609929078014,
|
| 1157 |
+
"grad_norm": 0.7693955302238464,
|
| 1158 |
+
"learning_rate": 2.4231678486997638e-06,
|
| 1159 |
+
"loss": 0.1303,
|
| 1160 |
+
"step": 1610
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 2.872340425531915,
|
| 1164 |
+
"grad_norm": 3.25712251663208,
|
| 1165 |
+
"learning_rate": 2.1276595744680853e-06,
|
| 1166 |
+
"loss": 0.0748,
|
| 1167 |
+
"step": 1620
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 2.8900709219858154,
|
| 1171 |
+
"grad_norm": 3.0831634998321533,
|
| 1172 |
+
"learning_rate": 1.8321513002364066e-06,
|
| 1173 |
+
"loss": 0.0703,
|
| 1174 |
+
"step": 1630
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 2.9078014184397163,
|
| 1178 |
+
"grad_norm": 2.351789712905884,
|
| 1179 |
+
"learning_rate": 1.5366430260047282e-06,
|
| 1180 |
+
"loss": 0.0501,
|
| 1181 |
+
"step": 1640
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 2.925531914893617,
|
| 1185 |
+
"grad_norm": 2.154916286468506,
|
| 1186 |
+
"learning_rate": 1.2411347517730497e-06,
|
| 1187 |
+
"loss": 0.0907,
|
| 1188 |
+
"step": 1650
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 2.9432624113475176,
|
| 1192 |
+
"grad_norm": 6.570488452911377,
|
| 1193 |
+
"learning_rate": 9.456264775413712e-07,
|
| 1194 |
+
"loss": 0.1337,
|
| 1195 |
+
"step": 1660
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 2.9609929078014185,
|
| 1199 |
+
"grad_norm": 4.879380702972412,
|
| 1200 |
+
"learning_rate": 6.501182033096927e-07,
|
| 1201 |
+
"loss": 0.1201,
|
| 1202 |
+
"step": 1670
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 2.978723404255319,
|
| 1206 |
+
"grad_norm": 4.488287448883057,
|
| 1207 |
+
"learning_rate": 3.546099290780142e-07,
|
| 1208 |
+
"loss": 0.1064,
|
| 1209 |
+
"step": 1680
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 2.99645390070922,
|
| 1213 |
+
"grad_norm": 3.126821279525757,
|
| 1214 |
+
"learning_rate": 5.91016548463357e-08,
|
| 1215 |
+
"loss": 0.1375,
|
| 1216 |
+
"step": 1690
|
| 1217 |
+
}
|
| 1218 |
+
],
|
| 1219 |
+
"logging_steps": 10,
|
| 1220 |
+
"max_steps": 1692,
|
| 1221 |
+
"num_input_tokens_seen": 0,
|
| 1222 |
+
"num_train_epochs": 3,
|
| 1223 |
+
"save_steps": 500,
|
| 1224 |
+
"stateful_callbacks": {
|
| 1225 |
+
"TrainerControl": {
|
| 1226 |
+
"args": {
|
| 1227 |
+
"should_epoch_stop": false,
|
| 1228 |
+
"should_evaluate": false,
|
| 1229 |
+
"should_log": false,
|
| 1230 |
+
"should_save": true,
|
| 1231 |
+
"should_training_stop": true
|
| 1232 |
+
},
|
| 1233 |
+
"attributes": {}
|
| 1234 |
+
}
|
| 1235 |
+
},
|
| 1236 |
+
"total_flos": 7160790169147392.0,
|
| 1237 |
+
"train_batch_size": 32,
|
| 1238 |
+
"trial_name": null,
|
| 1239 |
+
"trial_params": null
|
| 1240 |
+
}
|
checkpoint-1692/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:589953756b2200bee9b4dd81fcad8270e00bca90c5da9b4f312663d0f0c1fc8c
|
| 3 |
+
size 5368
|
checkpoint-1692/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": "distilbert-base-uncased",
|
| 3 |
+
"activation": "gelu",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"DistilBertForSequenceClassification"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.1,
|
| 8 |
+
"dim": 768,
|
| 9 |
+
"dropout": 0.1,
|
| 10 |
+
"hidden_dim": 3072,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"max_position_embeddings": 512,
|
| 13 |
+
"model_type": "distilbert",
|
| 14 |
+
"n_heads": 12,
|
| 15 |
+
"n_layers": 6,
|
| 16 |
+
"pad_token_id": 0,
|
| 17 |
+
"problem_type": "single_label_classification",
|
| 18 |
+
"qa_dropout": 0.1,
|
| 19 |
+
"seq_classif_dropout": 0.2,
|
| 20 |
+
"sinusoidal_pos_embds": false,
|
| 21 |
+
"tie_weights_": true,
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.48.0",
|
| 24 |
+
"vocab_size": 30522
|
| 25 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b232355fadb9bbfb707eed2b6980060355c8d8c7e4b5dc79d821f4dd37af8d23
|
| 3 |
+
size 267832560
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
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 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 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": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 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": "DistilBertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:589953756b2200bee9b4dd81fcad8270e00bca90c5da9b4f312663d0f0c1fc8c
|
| 3 |
+
size 5368
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|