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": 99.8648280558393,
|
| 3 |
+
"total_flos": 3.848614248660486e+18,
|
| 4 |
+
"train_loss": 1.9516082713779377,
|
| 5 |
+
"train_runtime": 79711.1977,
|
| 6 |
+
"train_samples_per_second": 184.18,
|
| 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": 99.8648280558393,
|
| 3 |
+
"total_flos": 3.848614248660486e+18,
|
| 4 |
+
"train_loss": 1.9516082713779377,
|
| 5 |
+
"train_runtime": 79711.1977,
|
| 6 |
+
"train_samples_per_second": 184.18,
|
| 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": 99.8648280558393,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 73400,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.9996595165134491,
|
| 14 |
+
"grad_norm": 1.3896363973617554,
|
| 15 |
+
"learning_rate": 9.90040871934605e-06,
|
| 16 |
+
"loss": 2.3637,
|
| 17 |
+
"step": 734
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.9982975825672455,
|
| 21 |
+
"grad_norm": 1.085627794265747,
|
| 22 |
+
"learning_rate": 9.80040871934605e-06,
|
| 23 |
+
"loss": 2.3048,
|
| 24 |
+
"step": 1468
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 2.996935648621042,
|
| 28 |
+
"grad_norm": 1.7355504035949707,
|
| 29 |
+
"learning_rate": 9.700544959128066e-06,
|
| 30 |
+
"loss": 2.2626,
|
| 31 |
+
"step": 2202
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 3.9955737146748382,
|
| 35 |
+
"grad_norm": 2.631619691848755,
|
| 36 |
+
"learning_rate": 9.600544959128067e-06,
|
| 37 |
+
"loss": 2.2366,
|
| 38 |
+
"step": 2936
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.994211780728635,
|
| 42 |
+
"grad_norm": 3.0078556537628174,
|
| 43 |
+
"learning_rate": 9.500544959128066e-06,
|
| 44 |
+
"loss": 2.1902,
|
| 45 |
+
"step": 3670
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 5.992849846782431,
|
| 49 |
+
"grad_norm": 3.808694362640381,
|
| 50 |
+
"learning_rate": 9.400681198910083e-06,
|
| 51 |
+
"loss": 2.1761,
|
| 52 |
+
"step": 4404
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 6.991487912836227,
|
| 56 |
+
"grad_norm": 2.927497625350952,
|
| 57 |
+
"learning_rate": 9.300681198910082e-06,
|
| 58 |
+
"loss": 2.1423,
|
| 59 |
+
"step": 5138
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 7.990125978890024,
|
| 63 |
+
"grad_norm": 3.1034231185913086,
|
| 64 |
+
"learning_rate": 9.200681198910083e-06,
|
| 65 |
+
"loss": 2.1311,
|
| 66 |
+
"step": 5872
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 8.98876404494382,
|
| 70 |
+
"grad_norm": 2.8936243057250977,
|
| 71 |
+
"learning_rate": 9.1008174386921e-06,
|
| 72 |
+
"loss": 2.1006,
|
| 73 |
+
"step": 6606
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 9.987402110997616,
|
| 77 |
+
"grad_norm": 3.1552979946136475,
|
| 78 |
+
"learning_rate": 9.000817438692099e-06,
|
| 79 |
+
"loss": 2.1108,
|
| 80 |
+
"step": 7340
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 10.986040177051413,
|
| 84 |
+
"grad_norm": 4.212212562561035,
|
| 85 |
+
"learning_rate": 8.900817438692098e-06,
|
| 86 |
+
"loss": 2.0826,
|
| 87 |
+
"step": 8074
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 11.98467824310521,
|
| 91 |
+
"grad_norm": 3.0641448497772217,
|
| 92 |
+
"learning_rate": 8.800953678474115e-06,
|
| 93 |
+
"loss": 2.0827,
|
| 94 |
+
"step": 8808
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 12.983316309159006,
|
| 98 |
+
"grad_norm": 3.0309231281280518,
|
| 99 |
+
"learning_rate": 8.700953678474116e-06,
|
| 100 |
+
"loss": 2.0707,
|
| 101 |
+
"step": 9542
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 13.981954375212801,
|
| 105 |
+
"grad_norm": 2.910205364227295,
|
| 106 |
+
"learning_rate": 8.600953678474115e-06,
|
| 107 |
+
"loss": 2.0545,
|
| 108 |
+
"step": 10276
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 14.980592441266598,
|
| 112 |
+
"grad_norm": 2.4057796001434326,
|
| 113 |
+
"learning_rate": 8.501089918256132e-06,
|
| 114 |
+
"loss": 2.0518,
|
| 115 |
+
"step": 11010
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 15.979230507320395,
|
| 119 |
+
"grad_norm": 3.38080096244812,
|
| 120 |
+
"learning_rate": 8.401089918256131e-06,
|
| 121 |
+
"loss": 2.0431,
|
| 122 |
+
"step": 11744
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 16.97786857337419,
|
| 126 |
+
"grad_norm": 3.414673328399658,
|
| 127 |
+
"learning_rate": 8.301089918256132e-06,
|
| 128 |
+
"loss": 2.0278,
|
| 129 |
+
"step": 12478
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 17.97650663942799,
|
| 133 |
+
"grad_norm": 3.2804346084594727,
|
| 134 |
+
"learning_rate": 8.201089918256131e-06,
|
| 135 |
+
"loss": 2.0394,
|
| 136 |
+
"step": 13212
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 18.975144705481785,
|
| 140 |
+
"grad_norm": 3.313039541244507,
|
| 141 |
+
"learning_rate": 8.101089918256132e-06,
|
| 142 |
+
"loss": 2.0239,
|
| 143 |
+
"step": 13946
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 19.973782771535582,
|
| 147 |
+
"grad_norm": 3.2474405765533447,
|
| 148 |
+
"learning_rate": 8.001226158038149e-06,
|
| 149 |
+
"loss": 2.0155,
|
| 150 |
+
"step": 14680
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 20.972420837589375,
|
| 154 |
+
"grad_norm": 3.154122829437256,
|
| 155 |
+
"learning_rate": 7.901226158038148e-06,
|
| 156 |
+
"loss": 2.0183,
|
| 157 |
+
"step": 15414
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 21.971058903643172,
|
| 161 |
+
"grad_norm": 4.2719407081604,
|
| 162 |
+
"learning_rate": 7.801226158038147e-06,
|
| 163 |
+
"loss": 2.0059,
|
| 164 |
+
"step": 16148
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 22.96969696969697,
|
| 168 |
+
"grad_norm": 3.08585262298584,
|
| 169 |
+
"learning_rate": 7.701226158038148e-06,
|
| 170 |
+
"loss": 2.007,
|
| 171 |
+
"step": 16882
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 23.968335035750766,
|
| 175 |
+
"grad_norm": 3.8600127696990967,
|
| 176 |
+
"learning_rate": 7.601362397820165e-06,
|
| 177 |
+
"loss": 1.9813,
|
| 178 |
+
"step": 17616
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 24.966973101804562,
|
| 182 |
+
"grad_norm": 4.064390182495117,
|
| 183 |
+
"learning_rate": 7.501362397820165e-06,
|
| 184 |
+
"loss": 1.9909,
|
| 185 |
+
"step": 18350
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 25.96561116785836,
|
| 189 |
+
"grad_norm": 5.1680169105529785,
|
| 190 |
+
"learning_rate": 7.401362397820164e-06,
|
| 191 |
+
"loss": 1.9781,
|
| 192 |
+
"step": 19084
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 26.964249233912156,
|
| 196 |
+
"grad_norm": 3.716388702392578,
|
| 197 |
+
"learning_rate": 7.30149863760218e-06,
|
| 198 |
+
"loss": 1.9841,
|
| 199 |
+
"step": 19818
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 27.962887299965953,
|
| 203 |
+
"grad_norm": 1.9636414051055908,
|
| 204 |
+
"learning_rate": 7.20149863760218e-06,
|
| 205 |
+
"loss": 1.9727,
|
| 206 |
+
"step": 20552
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 28.96152536601975,
|
| 210 |
+
"grad_norm": 2.968393087387085,
|
| 211 |
+
"learning_rate": 7.10149863760218e-06,
|
| 212 |
+
"loss": 1.9729,
|
| 213 |
+
"step": 21286
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 29.960163432073543,
|
| 217 |
+
"grad_norm": 1.9185447692871094,
|
| 218 |
+
"learning_rate": 7.001498637602181e-06,
|
| 219 |
+
"loss": 1.9586,
|
| 220 |
+
"step": 22020
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 30.95880149812734,
|
| 224 |
+
"grad_norm": 3.3125815391540527,
|
| 225 |
+
"learning_rate": 6.901634877384197e-06,
|
| 226 |
+
"loss": 1.9661,
|
| 227 |
+
"step": 22754
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 31.957439564181136,
|
| 231 |
+
"grad_norm": 3.3222153186798096,
|
| 232 |
+
"learning_rate": 6.801634877384197e-06,
|
| 233 |
+
"loss": 1.9532,
|
| 234 |
+
"step": 23488
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 32.95607763023494,
|
| 238 |
+
"grad_norm": 2.3255455493927,
|
| 239 |
+
"learning_rate": 6.701634877384197e-06,
|
| 240 |
+
"loss": 1.9445,
|
| 241 |
+
"step": 24222
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 33.95471569628873,
|
| 245 |
+
"grad_norm": 3.436879873275757,
|
| 246 |
+
"learning_rate": 6.6017711171662135e-06,
|
| 247 |
+
"loss": 1.9564,
|
| 248 |
+
"step": 24956
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 34.95335376234252,
|
| 252 |
+
"grad_norm": 2.789512872695923,
|
| 253 |
+
"learning_rate": 6.501771117166214e-06,
|
| 254 |
+
"loss": 1.9458,
|
| 255 |
+
"step": 25690
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 35.951991828396324,
|
| 259 |
+
"grad_norm": 2.496615409851074,
|
| 260 |
+
"learning_rate": 6.401771117166214e-06,
|
| 261 |
+
"loss": 1.9314,
|
| 262 |
+
"step": 26424
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 36.95062989445012,
|
| 266 |
+
"grad_norm": 3.680979013442993,
|
| 267 |
+
"learning_rate": 6.301907356948229e-06,
|
| 268 |
+
"loss": 1.9481,
|
| 269 |
+
"step": 27158
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 37.94926796050392,
|
| 273 |
+
"grad_norm": 3.8750393390655518,
|
| 274 |
+
"learning_rate": 6.201907356948229e-06,
|
| 275 |
+
"loss": 1.9346,
|
| 276 |
+
"step": 27892
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 38.94790602655771,
|
| 280 |
+
"grad_norm": 3.358818531036377,
|
| 281 |
+
"learning_rate": 6.101907356948229e-06,
|
| 282 |
+
"loss": 1.9426,
|
| 283 |
+
"step": 28626
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 39.94654409261151,
|
| 287 |
+
"grad_norm": 3.4985063076019287,
|
| 288 |
+
"learning_rate": 6.001907356948229e-06,
|
| 289 |
+
"loss": 1.9319,
|
| 290 |
+
"step": 29360
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 40.945182158665304,
|
| 294 |
+
"grad_norm": 3.5801708698272705,
|
| 295 |
+
"learning_rate": 5.9020435967302455e-06,
|
| 296 |
+
"loss": 1.9316,
|
| 297 |
+
"step": 30094
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 41.943820224719104,
|
| 301 |
+
"grad_norm": 2.131335973739624,
|
| 302 |
+
"learning_rate": 5.802043596730246e-06,
|
| 303 |
+
"loss": 1.9267,
|
| 304 |
+
"step": 30828
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 42.9424582907729,
|
| 308 |
+
"grad_norm": 3.048663854598999,
|
| 309 |
+
"learning_rate": 5.702043596730246e-06,
|
| 310 |
+
"loss": 1.9313,
|
| 311 |
+
"step": 31562
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 43.94109635682669,
|
| 315 |
+
"grad_norm": 4.18536376953125,
|
| 316 |
+
"learning_rate": 5.602043596730246e-06,
|
| 317 |
+
"loss": 1.9307,
|
| 318 |
+
"step": 32296
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 44.93973442288049,
|
| 322 |
+
"grad_norm": 3.1449191570281982,
|
| 323 |
+
"learning_rate": 5.502043596730245e-06,
|
| 324 |
+
"loss": 1.9121,
|
| 325 |
+
"step": 33030
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 45.938372488934284,
|
| 329 |
+
"grad_norm": 2.445446491241455,
|
| 330 |
+
"learning_rate": 5.4021798365122625e-06,
|
| 331 |
+
"loss": 1.9147,
|
| 332 |
+
"step": 33764
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 46.937010554988085,
|
| 336 |
+
"grad_norm": 3.044128656387329,
|
| 337 |
+
"learning_rate": 5.302316076294278e-06,
|
| 338 |
+
"loss": 1.9179,
|
| 339 |
+
"step": 34498
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 47.93564862104188,
|
| 343 |
+
"grad_norm": 2.411952018737793,
|
| 344 |
+
"learning_rate": 5.202316076294278e-06,
|
| 345 |
+
"loss": 1.9175,
|
| 346 |
+
"step": 35232
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 48.93428668709568,
|
| 350 |
+
"grad_norm": 4.5417799949646,
|
| 351 |
+
"learning_rate": 5.1023160762942784e-06,
|
| 352 |
+
"loss": 1.9139,
|
| 353 |
+
"step": 35966
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 49.93292475314947,
|
| 357 |
+
"grad_norm": 3.36789608001709,
|
| 358 |
+
"learning_rate": 5.002452316076295e-06,
|
| 359 |
+
"loss": 1.8999,
|
| 360 |
+
"step": 36700
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 50.93156281920327,
|
| 364 |
+
"grad_norm": 1.4864888191223145,
|
| 365 |
+
"learning_rate": 4.902452316076295e-06,
|
| 366 |
+
"loss": 1.9151,
|
| 367 |
+
"step": 37434
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 51.930200885257065,
|
| 371 |
+
"grad_norm": 2.543835163116455,
|
| 372 |
+
"learning_rate": 4.802452316076295e-06,
|
| 373 |
+
"loss": 1.9049,
|
| 374 |
+
"step": 38168
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 52.92883895131086,
|
| 378 |
+
"grad_norm": 2.121729850769043,
|
| 379 |
+
"learning_rate": 4.7024523160762945e-06,
|
| 380 |
+
"loss": 1.9099,
|
| 381 |
+
"step": 38902
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 53.92747701736466,
|
| 385 |
+
"grad_norm": 3.2268223762512207,
|
| 386 |
+
"learning_rate": 4.602452316076295e-06,
|
| 387 |
+
"loss": 1.9071,
|
| 388 |
+
"step": 39636
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 54.92611508341845,
|
| 392 |
+
"grad_norm": 3.928720235824585,
|
| 393 |
+
"learning_rate": 4.502452316076295e-06,
|
| 394 |
+
"loss": 1.8883,
|
| 395 |
+
"step": 40370
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 55.92475314947225,
|
| 399 |
+
"grad_norm": 2.6059372425079346,
|
| 400 |
+
"learning_rate": 4.402588555858311e-06,
|
| 401 |
+
"loss": 1.9033,
|
| 402 |
+
"step": 41104
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 56.923391215526046,
|
| 406 |
+
"grad_norm": 2.899505615234375,
|
| 407 |
+
"learning_rate": 4.3025885558583105e-06,
|
| 408 |
+
"loss": 1.8852,
|
| 409 |
+
"step": 41838
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 57.922029281579846,
|
| 413 |
+
"grad_norm": 2.1250553131103516,
|
| 414 |
+
"learning_rate": 4.202588555858311e-06,
|
| 415 |
+
"loss": 1.8919,
|
| 416 |
+
"step": 42572
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 58.92066734763364,
|
| 420 |
+
"grad_norm": 3.6068265438079834,
|
| 421 |
+
"learning_rate": 4.102588555858311e-06,
|
| 422 |
+
"loss": 1.8801,
|
| 423 |
+
"step": 43306
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 59.91930541368744,
|
| 427 |
+
"grad_norm": 3.1520488262176514,
|
| 428 |
+
"learning_rate": 4.002588555858311e-06,
|
| 429 |
+
"loss": 1.8997,
|
| 430 |
+
"step": 44040
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 60.91794347974123,
|
| 434 |
+
"grad_norm": 2.7997710704803467,
|
| 435 |
+
"learning_rate": 3.902588555858311e-06,
|
| 436 |
+
"loss": 1.8942,
|
| 437 |
+
"step": 44774
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 61.916581545795026,
|
| 441 |
+
"grad_norm": 2.2547781467437744,
|
| 442 |
+
"learning_rate": 3.8027247956403276e-06,
|
| 443 |
+
"loss": 1.8996,
|
| 444 |
+
"step": 45508
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 62.915219611848826,
|
| 448 |
+
"grad_norm": 2.583367347717285,
|
| 449 |
+
"learning_rate": 3.7027247956403272e-06,
|
| 450 |
+
"loss": 1.8811,
|
| 451 |
+
"step": 46242
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 63.91385767790262,
|
| 455 |
+
"grad_norm": 3.9086215496063232,
|
| 456 |
+
"learning_rate": 3.6027247956403274e-06,
|
| 457 |
+
"loss": 1.8878,
|
| 458 |
+
"step": 46976
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 64.91249574395641,
|
| 462 |
+
"grad_norm": 1.8989681005477905,
|
| 463 |
+
"learning_rate": 3.50299727520436e-06,
|
| 464 |
+
"loss": 1.8987,
|
| 465 |
+
"step": 47710
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 65.91113381001021,
|
| 469 |
+
"grad_norm": 2.5425479412078857,
|
| 470 |
+
"learning_rate": 3.40299727520436e-06,
|
| 471 |
+
"loss": 1.8928,
|
| 472 |
+
"step": 48444
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 66.90977187606401,
|
| 476 |
+
"grad_norm": 2.8161380290985107,
|
| 477 |
+
"learning_rate": 3.3029972752043598e-06,
|
| 478 |
+
"loss": 1.8833,
|
| 479 |
+
"step": 49178
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 67.90840994211781,
|
| 483 |
+
"grad_norm": 3.328853130340576,
|
| 484 |
+
"learning_rate": 3.20299727520436e-06,
|
| 485 |
+
"loss": 1.8802,
|
| 486 |
+
"step": 49912
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 68.9070480081716,
|
| 490 |
+
"grad_norm": 1.836472749710083,
|
| 491 |
+
"learning_rate": 3.1029972752043596e-06,
|
| 492 |
+
"loss": 1.8821,
|
| 493 |
+
"step": 50646
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 69.9056860742254,
|
| 497 |
+
"grad_norm": 3.308563470840454,
|
| 498 |
+
"learning_rate": 3.00299727520436e-06,
|
| 499 |
+
"loss": 1.8824,
|
| 500 |
+
"step": 51380
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 70.9043241402792,
|
| 504 |
+
"grad_norm": 2.7526936531066895,
|
| 505 |
+
"learning_rate": 2.9031335149863767e-06,
|
| 506 |
+
"loss": 1.8828,
|
| 507 |
+
"step": 52114
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 71.90296220633299,
|
| 511 |
+
"grad_norm": 2.601332664489746,
|
| 512 |
+
"learning_rate": 2.8031335149863763e-06,
|
| 513 |
+
"loss": 1.8872,
|
| 514 |
+
"step": 52848
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 72.90160027238679,
|
| 518 |
+
"grad_norm": 3.038666248321533,
|
| 519 |
+
"learning_rate": 2.7031335149863765e-06,
|
| 520 |
+
"loss": 1.8861,
|
| 521 |
+
"step": 53582
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 73.90023833844059,
|
| 525 |
+
"grad_norm": 2.3815577030181885,
|
| 526 |
+
"learning_rate": 2.6032697547683926e-06,
|
| 527 |
+
"loss": 1.8768,
|
| 528 |
+
"step": 54316
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 74.89887640449439,
|
| 532 |
+
"grad_norm": 3.253783941268921,
|
| 533 |
+
"learning_rate": 2.5032697547683927e-06,
|
| 534 |
+
"loss": 1.8767,
|
| 535 |
+
"step": 55050
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 75.89751447054817,
|
| 539 |
+
"grad_norm": 1.8898478746414185,
|
| 540 |
+
"learning_rate": 2.403405994550409e-06,
|
| 541 |
+
"loss": 1.8778,
|
| 542 |
+
"step": 55784
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 76.89615253660197,
|
| 546 |
+
"grad_norm": 2.8832762241363525,
|
| 547 |
+
"learning_rate": 2.303405994550409e-06,
|
| 548 |
+
"loss": 1.8816,
|
| 549 |
+
"step": 56518
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 77.89479060265577,
|
| 553 |
+
"grad_norm": 2.767455577850342,
|
| 554 |
+
"learning_rate": 2.2034059945504087e-06,
|
| 555 |
+
"loss": 1.8817,
|
| 556 |
+
"step": 57252
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 78.89342866870956,
|
| 560 |
+
"grad_norm": 1.9658331871032715,
|
| 561 |
+
"learning_rate": 2.1035422343324252e-06,
|
| 562 |
+
"loss": 1.8857,
|
| 563 |
+
"step": 57986
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 79.89206673476336,
|
| 567 |
+
"grad_norm": 2.7493135929107666,
|
| 568 |
+
"learning_rate": 2.003542234332425e-06,
|
| 569 |
+
"loss": 1.8763,
|
| 570 |
+
"step": 58720
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 80.89070480081716,
|
| 574 |
+
"grad_norm": 2.519388198852539,
|
| 575 |
+
"learning_rate": 1.9035422343324252e-06,
|
| 576 |
+
"loss": 1.8782,
|
| 577 |
+
"step": 59454
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 81.88934286687096,
|
| 581 |
+
"grad_norm": 2.258284091949463,
|
| 582 |
+
"learning_rate": 1.8036784741144416e-06,
|
| 583 |
+
"loss": 1.8709,
|
| 584 |
+
"step": 60188
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 82.88798093292475,
|
| 588 |
+
"grad_norm": 2.6191112995147705,
|
| 589 |
+
"learning_rate": 1.7038147138964577e-06,
|
| 590 |
+
"loss": 1.8486,
|
| 591 |
+
"step": 60922
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 83.88661899897855,
|
| 595 |
+
"grad_norm": 3.241089105606079,
|
| 596 |
+
"learning_rate": 1.603814713896458e-06,
|
| 597 |
+
"loss": 1.8726,
|
| 598 |
+
"step": 61656
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 84.88525706503235,
|
| 602 |
+
"grad_norm": 1.7192323207855225,
|
| 603 |
+
"learning_rate": 1.503814713896458e-06,
|
| 604 |
+
"loss": 1.8793,
|
| 605 |
+
"step": 62390
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 85.88389513108615,
|
| 609 |
+
"grad_norm": 2.782193899154663,
|
| 610 |
+
"learning_rate": 1.4038147138964579e-06,
|
| 611 |
+
"loss": 1.8714,
|
| 612 |
+
"step": 63124
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 86.88253319713994,
|
| 616 |
+
"grad_norm": 2.659689426422119,
|
| 617 |
+
"learning_rate": 1.303814713896458e-06,
|
| 618 |
+
"loss": 1.8908,
|
| 619 |
+
"step": 63858
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 87.88117126319374,
|
| 623 |
+
"grad_norm": 1.9582741260528564,
|
| 624 |
+
"learning_rate": 1.2038147138964579e-06,
|
| 625 |
+
"loss": 1.8595,
|
| 626 |
+
"step": 64592
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 88.87980932924754,
|
| 630 |
+
"grad_norm": 1.9837833642959595,
|
| 631 |
+
"learning_rate": 1.1039509536784742e-06,
|
| 632 |
+
"loss": 1.875,
|
| 633 |
+
"step": 65326
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 89.87844739530132,
|
| 637 |
+
"grad_norm": 2.076040506362915,
|
| 638 |
+
"learning_rate": 1.0039509536784741e-06,
|
| 639 |
+
"loss": 1.8664,
|
| 640 |
+
"step": 66060
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 90.87708546135512,
|
| 644 |
+
"grad_norm": 1.5430365800857544,
|
| 645 |
+
"learning_rate": 9.039509536784742e-07,
|
| 646 |
+
"loss": 1.8875,
|
| 647 |
+
"step": 66794
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 91.87572352740892,
|
| 651 |
+
"grad_norm": 1.8234200477600098,
|
| 652 |
+
"learning_rate": 8.040871934604905e-07,
|
| 653 |
+
"loss": 1.868,
|
| 654 |
+
"step": 67528
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 92.87436159346272,
|
| 658 |
+
"grad_norm": 2.1992905139923096,
|
| 659 |
+
"learning_rate": 7.040871934604905e-07,
|
| 660 |
+
"loss": 1.8844,
|
| 661 |
+
"step": 68262
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 93.87299965951651,
|
| 665 |
+
"grad_norm": 2.1095755100250244,
|
| 666 |
+
"learning_rate": 6.042234332425069e-07,
|
| 667 |
+
"loss": 1.8808,
|
| 668 |
+
"step": 68996
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 94.87163772557031,
|
| 672 |
+
"grad_norm": 2.1402716636657715,
|
| 673 |
+
"learning_rate": 5.042234332425069e-07,
|
| 674 |
+
"loss": 1.8661,
|
| 675 |
+
"step": 69730
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 95.87027579162411,
|
| 679 |
+
"grad_norm": 2.9710769653320312,
|
| 680 |
+
"learning_rate": 4.0422343324250687e-07,
|
| 681 |
+
"loss": 1.8651,
|
| 682 |
+
"step": 70464
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 96.8689138576779,
|
| 686 |
+
"grad_norm": 1.769300103187561,
|
| 687 |
+
"learning_rate": 3.043596730245232e-07,
|
| 688 |
+
"loss": 1.8726,
|
| 689 |
+
"step": 71198
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 97.8675519237317,
|
| 693 |
+
"grad_norm": 2.095330238342285,
|
| 694 |
+
"learning_rate": 2.0435967302452318e-07,
|
| 695 |
+
"loss": 1.8691,
|
| 696 |
+
"step": 71932
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 98.8661899897855,
|
| 700 |
+
"grad_norm": 2.741286039352417,
|
| 701 |
+
"learning_rate": 1.0435967302452316e-07,
|
| 702 |
+
"loss": 1.8809,
|
| 703 |
+
"step": 72666
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 99.8648280558393,
|
| 707 |
+
"grad_norm": 2.270244836807251,
|
| 708 |
+
"learning_rate": 4.49591280653951e-09,
|
| 709 |
+
"loss": 1.8619,
|
| 710 |
+
"step": 73400
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 99.8648280558393,
|
| 714 |
+
"step": 73400,
|
| 715 |
+
"total_flos": 3.848614248660486e+18,
|
| 716 |
+
"train_loss": 1.9516082713779377,
|
| 717 |
+
"train_runtime": 79711.1977,
|
| 718 |
+
"train_samples_per_second": 184.18,
|
| 719 |
+
"train_steps_per_second": 0.921
|
| 720 |
+
}
|
| 721 |
+
],
|
| 722 |
+
"logging_steps": 734,
|
| 723 |
+
"max_steps": 73400,
|
| 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": 3.848614248660486e+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
|
|
|