End of training
Browse files- all_results.json +8 -0
- preprocessor_config.json +28 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +58 -0
- train_results.json +8 -0
- trainer_state.json +743 -0
- vocab.txt +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 99.48249027237354,
|
| 3 |
+
"total_flos": 1.0064407234989773e+18,
|
| 4 |
+
"train_loss": 2.0492228651046753,
|
| 5 |
+
"train_runtime": 20851.1432,
|
| 6 |
+
"train_samples_per_second": 184.834,
|
| 7 |
+
"train_steps_per_second": 0.921
|
| 8 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_size": {
|
| 3 |
+
"height": 224,
|
| 4 |
+
"width": 224
|
| 5 |
+
},
|
| 6 |
+
"do_center_crop": false,
|
| 7 |
+
"do_convert_rgb": true,
|
| 8 |
+
"do_normalize": true,
|
| 9 |
+
"do_rescale": true,
|
| 10 |
+
"do_resize": true,
|
| 11 |
+
"image_mean": [
|
| 12 |
+
0.48145466,
|
| 13 |
+
0.4578275,
|
| 14 |
+
0.40821073
|
| 15 |
+
],
|
| 16 |
+
"image_processor_type": "ChineseCLIPImageProcessor",
|
| 17 |
+
"image_std": [
|
| 18 |
+
0.26862954,
|
| 19 |
+
0.26130258,
|
| 20 |
+
0.27577711
|
| 21 |
+
],
|
| 22 |
+
"resample": 3,
|
| 23 |
+
"rescale_factor": 0.00392156862745098,
|
| 24 |
+
"size": {
|
| 25 |
+
"height": 224,
|
| 26 |
+
"width": 224
|
| 27 |
+
}
|
| 28 |
+
}
|
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 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "BertTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 99.48249027237354,
|
| 3 |
+
"total_flos": 1.0064407234989773e+18,
|
| 4 |
+
"train_loss": 2.0492228651046753,
|
| 5 |
+
"train_runtime": 20851.1432,
|
| 6 |
+
"train_samples_per_second": 184.834,
|
| 7 |
+
"train_steps_per_second": 0.921
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,743 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 99.48249027237354,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 19200,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.9961089494163424,
|
| 14 |
+
"grad_norm": 0.9873842597007751,
|
| 15 |
+
"learning_rate": 9.901562500000001e-06,
|
| 16 |
+
"loss": 2.5953,
|
| 17 |
+
"step": 192
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.9909208819714657,
|
| 21 |
+
"grad_norm": 1.439842700958252,
|
| 22 |
+
"learning_rate": 9.8015625e-06,
|
| 23 |
+
"loss": 2.462,
|
| 24 |
+
"step": 384
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 2.985732814526589,
|
| 28 |
+
"grad_norm": 2.307729482650757,
|
| 29 |
+
"learning_rate": 9.701562500000001e-06,
|
| 30 |
+
"loss": 2.3659,
|
| 31 |
+
"step": 576
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 3.980544747081712,
|
| 35 |
+
"grad_norm": 4.006823539733887,
|
| 36 |
+
"learning_rate": 9.6015625e-06,
|
| 37 |
+
"loss": 2.3475,
|
| 38 |
+
"step": 768
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.975356679636835,
|
| 42 |
+
"grad_norm": 2.971815824508667,
|
| 43 |
+
"learning_rate": 9.501562500000001e-06,
|
| 44 |
+
"loss": 2.3162,
|
| 45 |
+
"step": 960
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 5.970168612191959,
|
| 49 |
+
"grad_norm": 3.6012327671051025,
|
| 50 |
+
"learning_rate": 9.4015625e-06,
|
| 51 |
+
"loss": 2.3049,
|
| 52 |
+
"step": 1152
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 6.964980544747082,
|
| 56 |
+
"grad_norm": 3.803527355194092,
|
| 57 |
+
"learning_rate": 9.301562500000002e-06,
|
| 58 |
+
"loss": 2.2733,
|
| 59 |
+
"step": 1344
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 7.959792477302205,
|
| 63 |
+
"grad_norm": 3.0761234760284424,
|
| 64 |
+
"learning_rate": 9.201562500000001e-06,
|
| 65 |
+
"loss": 2.2576,
|
| 66 |
+
"step": 1536
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 8.954604409857328,
|
| 70 |
+
"grad_norm": 2.6708500385284424,
|
| 71 |
+
"learning_rate": 9.1015625e-06,
|
| 72 |
+
"loss": 2.2367,
|
| 73 |
+
"step": 1728
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 9.949416342412452,
|
| 77 |
+
"grad_norm": 3.482462167739868,
|
| 78 |
+
"learning_rate": 9.001562500000001e-06,
|
| 79 |
+
"loss": 2.176,
|
| 80 |
+
"step": 1920
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 10.944228274967575,
|
| 84 |
+
"grad_norm": 2.826169967651367,
|
| 85 |
+
"learning_rate": 8.902083333333334e-06,
|
| 86 |
+
"loss": 2.2051,
|
| 87 |
+
"step": 2112
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 11.939040207522698,
|
| 91 |
+
"grad_norm": 3.164243698120117,
|
| 92 |
+
"learning_rate": 8.802083333333335e-06,
|
| 93 |
+
"loss": 2.1874,
|
| 94 |
+
"step": 2304
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 12.933852140077821,
|
| 98 |
+
"grad_norm": 4.725254058837891,
|
| 99 |
+
"learning_rate": 8.702083333333334e-06,
|
| 100 |
+
"loss": 2.1809,
|
| 101 |
+
"step": 2496
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 13.928664072632944,
|
| 105 |
+
"grad_norm": 3.7806782722473145,
|
| 106 |
+
"learning_rate": 8.602083333333333e-06,
|
| 107 |
+
"loss": 2.16,
|
| 108 |
+
"step": 2688
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 14.923476005188068,
|
| 112 |
+
"grad_norm": 3.735746145248413,
|
| 113 |
+
"learning_rate": 8.502083333333334e-06,
|
| 114 |
+
"loss": 2.1663,
|
| 115 |
+
"step": 2880
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 15.918287937743191,
|
| 119 |
+
"grad_norm": 3.401533365249634,
|
| 120 |
+
"learning_rate": 8.402083333333334e-06,
|
| 121 |
+
"loss": 2.1353,
|
| 122 |
+
"step": 3072
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 16.913099870298314,
|
| 126 |
+
"grad_norm": 3.1076550483703613,
|
| 127 |
+
"learning_rate": 8.302083333333335e-06,
|
| 128 |
+
"loss": 2.1277,
|
| 129 |
+
"step": 3264
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 17.907911802853437,
|
| 133 |
+
"grad_norm": 3.2938179969787598,
|
| 134 |
+
"learning_rate": 8.202083333333334e-06,
|
| 135 |
+
"loss": 2.1165,
|
| 136 |
+
"step": 3456
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 18.90272373540856,
|
| 140 |
+
"grad_norm": 3.4992620944976807,
|
| 141 |
+
"learning_rate": 8.102083333333333e-06,
|
| 142 |
+
"loss": 2.1407,
|
| 143 |
+
"step": 3648
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 19.897535667963684,
|
| 147 |
+
"grad_norm": 5.144476413726807,
|
| 148 |
+
"learning_rate": 8.002083333333334e-06,
|
| 149 |
+
"loss": 2.1124,
|
| 150 |
+
"step": 3840
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 20.892347600518807,
|
| 154 |
+
"grad_norm": 3.369130849838257,
|
| 155 |
+
"learning_rate": 7.902083333333333e-06,
|
| 156 |
+
"loss": 2.0994,
|
| 157 |
+
"step": 4032
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 21.88715953307393,
|
| 161 |
+
"grad_norm": 2.5219101905822754,
|
| 162 |
+
"learning_rate": 7.802604166666668e-06,
|
| 163 |
+
"loss": 2.086,
|
| 164 |
+
"step": 4224
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 22.881971465629054,
|
| 168 |
+
"grad_norm": 2.416583776473999,
|
| 169 |
+
"learning_rate": 7.702604166666667e-06,
|
| 170 |
+
"loss": 2.1011,
|
| 171 |
+
"step": 4416
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 23.876783398184177,
|
| 175 |
+
"grad_norm": 2.880645513534546,
|
| 176 |
+
"learning_rate": 7.602604166666667e-06,
|
| 177 |
+
"loss": 2.069,
|
| 178 |
+
"step": 4608
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 24.8715953307393,
|
| 182 |
+
"grad_norm": 3.6908700466156006,
|
| 183 |
+
"learning_rate": 7.5026041666666675e-06,
|
| 184 |
+
"loss": 2.0878,
|
| 185 |
+
"step": 4800
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 25.866407263294423,
|
| 189 |
+
"grad_norm": 3.9834940433502197,
|
| 190 |
+
"learning_rate": 7.402604166666667e-06,
|
| 191 |
+
"loss": 2.0832,
|
| 192 |
+
"step": 4992
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 26.861219195849547,
|
| 196 |
+
"grad_norm": 4.434403896331787,
|
| 197 |
+
"learning_rate": 7.302604166666667e-06,
|
| 198 |
+
"loss": 2.0618,
|
| 199 |
+
"step": 5184
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 27.85603112840467,
|
| 203 |
+
"grad_norm": 2.6179683208465576,
|
| 204 |
+
"learning_rate": 7.202604166666667e-06,
|
| 205 |
+
"loss": 2.0698,
|
| 206 |
+
"step": 5376
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 28.850843060959793,
|
| 210 |
+
"grad_norm": 3.7101306915283203,
|
| 211 |
+
"learning_rate": 7.102604166666667e-06,
|
| 212 |
+
"loss": 2.0561,
|
| 213 |
+
"step": 5568
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 29.845654993514916,
|
| 217 |
+
"grad_norm": 2.940573215484619,
|
| 218 |
+
"learning_rate": 7.002604166666668e-06,
|
| 219 |
+
"loss": 2.0499,
|
| 220 |
+
"step": 5760
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 30.84046692607004,
|
| 224 |
+
"grad_norm": 3.1798958778381348,
|
| 225 |
+
"learning_rate": 6.902604166666667e-06,
|
| 226 |
+
"loss": 2.049,
|
| 227 |
+
"step": 5952
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 31.835278858625163,
|
| 231 |
+
"grad_norm": 3.288163423538208,
|
| 232 |
+
"learning_rate": 6.802604166666667e-06,
|
| 233 |
+
"loss": 2.0326,
|
| 234 |
+
"step": 6144
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 32.83009079118028,
|
| 238 |
+
"grad_norm": 2.8764078617095947,
|
| 239 |
+
"learning_rate": 6.703125e-06,
|
| 240 |
+
"loss": 2.0371,
|
| 241 |
+
"step": 6336
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 33.824902723735406,
|
| 245 |
+
"grad_norm": 2.9869441986083984,
|
| 246 |
+
"learning_rate": 6.603125e-06,
|
| 247 |
+
"loss": 2.0611,
|
| 248 |
+
"step": 6528
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 34.81971465629053,
|
| 252 |
+
"grad_norm": 3.1720468997955322,
|
| 253 |
+
"learning_rate": 6.5031250000000005e-06,
|
| 254 |
+
"loss": 2.0541,
|
| 255 |
+
"step": 6720
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 35.81452658884565,
|
| 259 |
+
"grad_norm": 3.7449281215667725,
|
| 260 |
+
"learning_rate": 6.403125e-06,
|
| 261 |
+
"loss": 2.0177,
|
| 262 |
+
"step": 6912
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 36.809338521400775,
|
| 266 |
+
"grad_norm": 2.036007881164551,
|
| 267 |
+
"learning_rate": 6.303125000000001e-06,
|
| 268 |
+
"loss": 2.006,
|
| 269 |
+
"step": 7104
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 37.8041504539559,
|
| 273 |
+
"grad_norm": 2.677340507507324,
|
| 274 |
+
"learning_rate": 6.203125000000001e-06,
|
| 275 |
+
"loss": 2.0329,
|
| 276 |
+
"step": 7296
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 38.79896238651102,
|
| 280 |
+
"grad_norm": 2.8975298404693604,
|
| 281 |
+
"learning_rate": 6.103125000000001e-06,
|
| 282 |
+
"loss": 2.0347,
|
| 283 |
+
"step": 7488
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 39.793774319066145,
|
| 287 |
+
"grad_norm": 1.8516769409179688,
|
| 288 |
+
"learning_rate": 6.003125000000001e-06,
|
| 289 |
+
"loss": 2.0179,
|
| 290 |
+
"step": 7680
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 40.78858625162127,
|
| 294 |
+
"grad_norm": 2.984468698501587,
|
| 295 |
+
"learning_rate": 5.903125e-06,
|
| 296 |
+
"loss": 2.0002,
|
| 297 |
+
"step": 7872
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 41.78339818417639,
|
| 301 |
+
"grad_norm": 4.378921031951904,
|
| 302 |
+
"learning_rate": 5.8031250000000004e-06,
|
| 303 |
+
"loss": 2.0032,
|
| 304 |
+
"step": 8064
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 42.778210116731515,
|
| 308 |
+
"grad_norm": 3.600853204727173,
|
| 309 |
+
"learning_rate": 5.7031250000000006e-06,
|
| 310 |
+
"loss": 2.0131,
|
| 311 |
+
"step": 8256
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 43.77302204928664,
|
| 315 |
+
"grad_norm": 3.475604295730591,
|
| 316 |
+
"learning_rate": 5.603125000000001e-06,
|
| 317 |
+
"loss": 2.0167,
|
| 318 |
+
"step": 8448
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 44.76783398184176,
|
| 322 |
+
"grad_norm": 3.7388381958007812,
|
| 323 |
+
"learning_rate": 5.503125e-06,
|
| 324 |
+
"loss": 2.0286,
|
| 325 |
+
"step": 8640
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 45.762645914396884,
|
| 329 |
+
"grad_norm": 3.0160775184631348,
|
| 330 |
+
"learning_rate": 5.403125e-06,
|
| 331 |
+
"loss": 2.0327,
|
| 332 |
+
"step": 8832
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 46.75745784695201,
|
| 336 |
+
"grad_norm": 2.4288928508758545,
|
| 337 |
+
"learning_rate": 5.303645833333334e-06,
|
| 338 |
+
"loss": 1.9837,
|
| 339 |
+
"step": 9024
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 47.75226977950713,
|
| 343 |
+
"grad_norm": 2.6072142124176025,
|
| 344 |
+
"learning_rate": 5.203645833333334e-06,
|
| 345 |
+
"loss": 2.022,
|
| 346 |
+
"step": 9216
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 48.747081712062254,
|
| 350 |
+
"grad_norm": 3.0784308910369873,
|
| 351 |
+
"learning_rate": 5.103645833333334e-06,
|
| 352 |
+
"loss": 1.9946,
|
| 353 |
+
"step": 9408
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 49.74189364461738,
|
| 357 |
+
"grad_norm": 2.552196979522705,
|
| 358 |
+
"learning_rate": 5.003645833333334e-06,
|
| 359 |
+
"loss": 1.9857,
|
| 360 |
+
"step": 9600
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 50.7367055771725,
|
| 364 |
+
"grad_norm": 5.163851261138916,
|
| 365 |
+
"learning_rate": 4.903645833333333e-06,
|
| 366 |
+
"loss": 2.0038,
|
| 367 |
+
"step": 9792
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 51.731517509727624,
|
| 371 |
+
"grad_norm": 3.9303905963897705,
|
| 372 |
+
"learning_rate": 4.8036458333333335e-06,
|
| 373 |
+
"loss": 2.0327,
|
| 374 |
+
"step": 9984
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 52.72632944228275,
|
| 378 |
+
"grad_norm": 4.269293308258057,
|
| 379 |
+
"learning_rate": 4.703645833333334e-06,
|
| 380 |
+
"loss": 1.978,
|
| 381 |
+
"step": 10176
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 53.72114137483787,
|
| 385 |
+
"grad_norm": 3.4755337238311768,
|
| 386 |
+
"learning_rate": 4.603645833333334e-06,
|
| 387 |
+
"loss": 2.0154,
|
| 388 |
+
"step": 10368
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 54.715953307392994,
|
| 392 |
+
"grad_norm": 2.2201080322265625,
|
| 393 |
+
"learning_rate": 4.503645833333334e-06,
|
| 394 |
+
"loss": 1.9945,
|
| 395 |
+
"step": 10560
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 55.71076523994812,
|
| 399 |
+
"grad_norm": 3.3768227100372314,
|
| 400 |
+
"learning_rate": 4.403645833333334e-06,
|
| 401 |
+
"loss": 1.9764,
|
| 402 |
+
"step": 10752
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 56.70557717250324,
|
| 406 |
+
"grad_norm": 2.9863104820251465,
|
| 407 |
+
"learning_rate": 4.303645833333334e-06,
|
| 408 |
+
"loss": 2.0031,
|
| 409 |
+
"step": 10944
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 57.70038910505836,
|
| 413 |
+
"grad_norm": 3.351330280303955,
|
| 414 |
+
"learning_rate": 4.203645833333333e-06,
|
| 415 |
+
"loss": 1.9915,
|
| 416 |
+
"step": 11136
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 58.69520103761349,
|
| 420 |
+
"grad_norm": 2.745793104171753,
|
| 421 |
+
"learning_rate": 4.103645833333333e-06,
|
| 422 |
+
"loss": 2.0054,
|
| 423 |
+
"step": 11328
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 59.69001297016861,
|
| 427 |
+
"grad_norm": 4.333703517913818,
|
| 428 |
+
"learning_rate": 4.0036458333333335e-06,
|
| 429 |
+
"loss": 1.9629,
|
| 430 |
+
"step": 11520
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 60.68482490272373,
|
| 434 |
+
"grad_norm": 3.1686012744903564,
|
| 435 |
+
"learning_rate": 3.903645833333334e-06,
|
| 436 |
+
"loss": 2.0011,
|
| 437 |
+
"step": 11712
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 61.679636835278856,
|
| 441 |
+
"grad_norm": 2.1179568767547607,
|
| 442 |
+
"learning_rate": 3.804166666666667e-06,
|
| 443 |
+
"loss": 1.9949,
|
| 444 |
+
"step": 11904
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 62.67444876783398,
|
| 448 |
+
"grad_norm": 2.4998276233673096,
|
| 449 |
+
"learning_rate": 3.704166666666667e-06,
|
| 450 |
+
"loss": 1.9791,
|
| 451 |
+
"step": 12096
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 63.6692607003891,
|
| 455 |
+
"grad_norm": 3.710357189178467,
|
| 456 |
+
"learning_rate": 3.6041666666666667e-06,
|
| 457 |
+
"loss": 1.9757,
|
| 458 |
+
"step": 12288
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 64.66407263294423,
|
| 462 |
+
"grad_norm": 3.4601991176605225,
|
| 463 |
+
"learning_rate": 3.504166666666667e-06,
|
| 464 |
+
"loss": 2.025,
|
| 465 |
+
"step": 12480
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 65.65888456549935,
|
| 469 |
+
"grad_norm": 2.626007318496704,
|
| 470 |
+
"learning_rate": 3.4041666666666665e-06,
|
| 471 |
+
"loss": 1.9755,
|
| 472 |
+
"step": 12672
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 66.65369649805447,
|
| 476 |
+
"grad_norm": 3.1454813480377197,
|
| 477 |
+
"learning_rate": 3.304166666666667e-06,
|
| 478 |
+
"loss": 1.9984,
|
| 479 |
+
"step": 12864
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 67.6485084306096,
|
| 483 |
+
"grad_norm": 7.283568859100342,
|
| 484 |
+
"learning_rate": 3.204166666666667e-06,
|
| 485 |
+
"loss": 1.9776,
|
| 486 |
+
"step": 13056
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 68.64332036316472,
|
| 490 |
+
"grad_norm": 2.031538724899292,
|
| 491 |
+
"learning_rate": 3.104166666666667e-06,
|
| 492 |
+
"loss": 1.9939,
|
| 493 |
+
"step": 13248
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 69.63813229571984,
|
| 497 |
+
"grad_norm": 2.370424270629883,
|
| 498 |
+
"learning_rate": 3.004166666666667e-06,
|
| 499 |
+
"loss": 1.9805,
|
| 500 |
+
"step": 13440
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 70.63294422827497,
|
| 504 |
+
"grad_norm": 3.6357874870300293,
|
| 505 |
+
"learning_rate": 2.9041666666666667e-06,
|
| 506 |
+
"loss": 1.9998,
|
| 507 |
+
"step": 13632
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 71.62775616083009,
|
| 511 |
+
"grad_norm": 3.61807918548584,
|
| 512 |
+
"learning_rate": 2.8041666666666668e-06,
|
| 513 |
+
"loss": 1.9615,
|
| 514 |
+
"step": 13824
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 72.62256809338521,
|
| 518 |
+
"grad_norm": 2.440490484237671,
|
| 519 |
+
"learning_rate": 2.7041666666666673e-06,
|
| 520 |
+
"loss": 1.9715,
|
| 521 |
+
"step": 14016
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 73.61738002594034,
|
| 525 |
+
"grad_norm": 2.490007162094116,
|
| 526 |
+
"learning_rate": 2.6046875000000006e-06,
|
| 527 |
+
"loss": 1.9765,
|
| 528 |
+
"step": 14208
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 74.61219195849546,
|
| 532 |
+
"grad_norm": 2.368825674057007,
|
| 533 |
+
"learning_rate": 2.5046875000000003e-06,
|
| 534 |
+
"loss": 1.9819,
|
| 535 |
+
"step": 14400
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 75.60700389105058,
|
| 539 |
+
"grad_norm": 2.9740967750549316,
|
| 540 |
+
"learning_rate": 2.4046875000000004e-06,
|
| 541 |
+
"loss": 1.9702,
|
| 542 |
+
"step": 14592
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 76.6018158236057,
|
| 546 |
+
"grad_norm": 3.215397834777832,
|
| 547 |
+
"learning_rate": 2.3046875e-06,
|
| 548 |
+
"loss": 1.9913,
|
| 549 |
+
"step": 14784
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 77.59662775616083,
|
| 553 |
+
"grad_norm": 2.465824604034424,
|
| 554 |
+
"learning_rate": 2.2046875000000002e-06,
|
| 555 |
+
"loss": 1.9751,
|
| 556 |
+
"step": 14976
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 78.59143968871595,
|
| 560 |
+
"grad_norm": 2.2337405681610107,
|
| 561 |
+
"learning_rate": 2.1046875000000003e-06,
|
| 562 |
+
"loss": 1.9496,
|
| 563 |
+
"step": 15168
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 79.58625162127107,
|
| 567 |
+
"grad_norm": 2.875598907470703,
|
| 568 |
+
"learning_rate": 2.0046875e-06,
|
| 569 |
+
"loss": 2.0066,
|
| 570 |
+
"step": 15360
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 80.5810635538262,
|
| 574 |
+
"grad_norm": 2.8990750312805176,
|
| 575 |
+
"learning_rate": 1.9046875000000001e-06,
|
| 576 |
+
"loss": 1.9466,
|
| 577 |
+
"step": 15552
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 81.57587548638132,
|
| 581 |
+
"grad_norm": 4.734499454498291,
|
| 582 |
+
"learning_rate": 1.8046875000000002e-06,
|
| 583 |
+
"loss": 1.9636,
|
| 584 |
+
"step": 15744
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 82.57068741893644,
|
| 588 |
+
"grad_norm": 2.9932515621185303,
|
| 589 |
+
"learning_rate": 1.7046875000000001e-06,
|
| 590 |
+
"loss": 1.973,
|
| 591 |
+
"step": 15936
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 83.56549935149157,
|
| 595 |
+
"grad_norm": 4.040909767150879,
|
| 596 |
+
"learning_rate": 1.6046875e-06,
|
| 597 |
+
"loss": 1.9678,
|
| 598 |
+
"step": 16128
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 84.56031128404669,
|
| 602 |
+
"grad_norm": 3.0035159587860107,
|
| 603 |
+
"learning_rate": 1.5046875000000002e-06,
|
| 604 |
+
"loss": 1.9457,
|
| 605 |
+
"step": 16320
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 85.55512321660181,
|
| 609 |
+
"grad_norm": 2.5469093322753906,
|
| 610 |
+
"learning_rate": 1.4046875e-06,
|
| 611 |
+
"loss": 1.9642,
|
| 612 |
+
"step": 16512
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 86.54993514915694,
|
| 616 |
+
"grad_norm": 2.850858688354492,
|
| 617 |
+
"learning_rate": 1.3046875e-06,
|
| 618 |
+
"loss": 1.9638,
|
| 619 |
+
"step": 16704
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 87.54474708171206,
|
| 623 |
+
"grad_norm": 2.5728836059570312,
|
| 624 |
+
"learning_rate": 1.2046875e-06,
|
| 625 |
+
"loss": 1.9681,
|
| 626 |
+
"step": 16896
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 88.53955901426718,
|
| 630 |
+
"grad_norm": 1.4205690622329712,
|
| 631 |
+
"learning_rate": 1.1046875000000002e-06,
|
| 632 |
+
"loss": 1.9585,
|
| 633 |
+
"step": 17088
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 89.5343709468223,
|
| 637 |
+
"grad_norm": 1.9591172933578491,
|
| 638 |
+
"learning_rate": 1.0046875e-06,
|
| 639 |
+
"loss": 1.9894,
|
| 640 |
+
"step": 17280
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 90.52918287937743,
|
| 644 |
+
"grad_norm": 3.981717348098755,
|
| 645 |
+
"learning_rate": 9.046875000000001e-07,
|
| 646 |
+
"loss": 1.9388,
|
| 647 |
+
"step": 17472
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 91.52399481193255,
|
| 651 |
+
"grad_norm": 2.348127841949463,
|
| 652 |
+
"learning_rate": 8.046875000000001e-07,
|
| 653 |
+
"loss": 1.9677,
|
| 654 |
+
"step": 17664
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 92.51880674448768,
|
| 658 |
+
"grad_norm": 3.0600669384002686,
|
| 659 |
+
"learning_rate": 7.046875e-07,
|
| 660 |
+
"loss": 1.9747,
|
| 661 |
+
"step": 17856
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 93.5136186770428,
|
| 665 |
+
"grad_norm": 3.865741491317749,
|
| 666 |
+
"learning_rate": 6.046875000000001e-07,
|
| 667 |
+
"loss": 1.958,
|
| 668 |
+
"step": 18048
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 94.50843060959792,
|
| 672 |
+
"grad_norm": 2.755720853805542,
|
| 673 |
+
"learning_rate": 5.052083333333334e-07,
|
| 674 |
+
"loss": 1.975,
|
| 675 |
+
"step": 18240
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 95.50324254215305,
|
| 679 |
+
"grad_norm": 3.4946985244750977,
|
| 680 |
+
"learning_rate": 4.0520833333333335e-07,
|
| 681 |
+
"loss": 1.9426,
|
| 682 |
+
"step": 18432
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 96.49805447470817,
|
| 686 |
+
"grad_norm": 2.993678331375122,
|
| 687 |
+
"learning_rate": 3.0520833333333336e-07,
|
| 688 |
+
"loss": 1.9497,
|
| 689 |
+
"step": 18624
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 97.49286640726329,
|
| 693 |
+
"grad_norm": 3.959327459335327,
|
| 694 |
+
"learning_rate": 2.0520833333333334e-07,
|
| 695 |
+
"loss": 1.9409,
|
| 696 |
+
"step": 18816
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 98.48767833981842,
|
| 700 |
+
"grad_norm": 2.147462844848633,
|
| 701 |
+
"learning_rate": 1.0520833333333334e-07,
|
| 702 |
+
"loss": 1.9485,
|
| 703 |
+
"step": 19008
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 99.48249027237354,
|
| 707 |
+
"grad_norm": 2.728996992111206,
|
| 708 |
+
"learning_rate": 5.208333333333334e-09,
|
| 709 |
+
"loss": 1.981,
|
| 710 |
+
"step": 19200
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 99.48249027237354,
|
| 714 |
+
"step": 19200,
|
| 715 |
+
"total_flos": 1.0064407234989773e+18,
|
| 716 |
+
"train_loss": 2.0492228651046753,
|
| 717 |
+
"train_runtime": 20851.1432,
|
| 718 |
+
"train_samples_per_second": 184.834,
|
| 719 |
+
"train_steps_per_second": 0.921
|
| 720 |
+
}
|
| 721 |
+
],
|
| 722 |
+
"logging_steps": 192,
|
| 723 |
+
"max_steps": 19200,
|
| 724 |
+
"num_input_tokens_seen": 0,
|
| 725 |
+
"num_train_epochs": 100,
|
| 726 |
+
"save_steps": 6800,
|
| 727 |
+
"stateful_callbacks": {
|
| 728 |
+
"TrainerControl": {
|
| 729 |
+
"args": {
|
| 730 |
+
"should_epoch_stop": false,
|
| 731 |
+
"should_evaluate": false,
|
| 732 |
+
"should_log": false,
|
| 733 |
+
"should_save": true,
|
| 734 |
+
"should_training_stop": true
|
| 735 |
+
},
|
| 736 |
+
"attributes": {}
|
| 737 |
+
}
|
| 738 |
+
},
|
| 739 |
+
"total_flos": 1.0064407234989773e+18,
|
| 740 |
+
"train_batch_size": 50,
|
| 741 |
+
"trial_name": null,
|
| 742 |
+
"trial_params": null
|
| 743 |
+
}
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|