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": 100.0,
|
| 3 |
+
"total_flos": 3.571154905966387e+18,
|
| 4 |
+
"train_loss": 1.9808363689442792,
|
| 5 |
+
"train_runtime": 84553.2908,
|
| 6 |
+
"train_samples_per_second": 160.897,
|
| 7 |
+
"train_steps_per_second": 1.341
|
| 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": 100.0,
|
| 3 |
+
"total_flos": 3.571154905966387e+18,
|
| 4 |
+
"train_loss": 1.9808363689442792,
|
| 5 |
+
"train_runtime": 84553.2908,
|
| 6 |
+
"train_samples_per_second": 160.897,
|
| 7 |
+
"train_steps_per_second": 1.341
|
| 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": 100.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 113400,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 1.0,
|
| 14 |
+
"grad_norm": 1.0368574857711792,
|
| 15 |
+
"learning_rate": 9.900440917107584e-06,
|
| 16 |
+
"loss": 2.4497,
|
| 17 |
+
"step": 1134
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 2.0,
|
| 21 |
+
"grad_norm": 2.962113380432129,
|
| 22 |
+
"learning_rate": 9.800440917107585e-06,
|
| 23 |
+
"loss": 2.3441,
|
| 24 |
+
"step": 2268
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 3.0,
|
| 28 |
+
"grad_norm": 2.04465913772583,
|
| 29 |
+
"learning_rate": 9.700529100529101e-06,
|
| 30 |
+
"loss": 2.2934,
|
| 31 |
+
"step": 3402
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 4.0,
|
| 35 |
+
"grad_norm": 3.710998773574829,
|
| 36 |
+
"learning_rate": 9.6005291005291e-06,
|
| 37 |
+
"loss": 2.2626,
|
| 38 |
+
"step": 4536
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 5.0,
|
| 42 |
+
"grad_norm": 2.217125177383423,
|
| 43 |
+
"learning_rate": 9.500529100529102e-06,
|
| 44 |
+
"loss": 2.2311,
|
| 45 |
+
"step": 5670
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 6.0,
|
| 49 |
+
"grad_norm": 4.2306671142578125,
|
| 50 |
+
"learning_rate": 9.400617283950619e-06,
|
| 51 |
+
"loss": 2.1809,
|
| 52 |
+
"step": 6804
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 7.0,
|
| 56 |
+
"grad_norm": 3.223266124725342,
|
| 57 |
+
"learning_rate": 9.300617283950618e-06,
|
| 58 |
+
"loss": 2.1686,
|
| 59 |
+
"step": 7938
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 8.0,
|
| 63 |
+
"grad_norm": 2.7807767391204834,
|
| 64 |
+
"learning_rate": 9.200705467372135e-06,
|
| 65 |
+
"loss": 2.139,
|
| 66 |
+
"step": 9072
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 9.0,
|
| 70 |
+
"grad_norm": 6.859907150268555,
|
| 71 |
+
"learning_rate": 9.100705467372136e-06,
|
| 72 |
+
"loss": 2.1258,
|
| 73 |
+
"step": 10206
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 10.0,
|
| 77 |
+
"grad_norm": 3.691579818725586,
|
| 78 |
+
"learning_rate": 9.000793650793651e-06,
|
| 79 |
+
"loss": 2.1096,
|
| 80 |
+
"step": 11340
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 11.0,
|
| 84 |
+
"grad_norm": 4.632028579711914,
|
| 85 |
+
"learning_rate": 8.90079365079365e-06,
|
| 86 |
+
"loss": 2.1051,
|
| 87 |
+
"step": 12474
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 12.0,
|
| 91 |
+
"grad_norm": 5.086824417114258,
|
| 92 |
+
"learning_rate": 8.800881834215167e-06,
|
| 93 |
+
"loss": 2.0945,
|
| 94 |
+
"step": 13608
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 13.0,
|
| 98 |
+
"grad_norm": 3.4331443309783936,
|
| 99 |
+
"learning_rate": 8.700881834215168e-06,
|
| 100 |
+
"loss": 2.0912,
|
| 101 |
+
"step": 14742
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 14.0,
|
| 105 |
+
"grad_norm": 4.4025092124938965,
|
| 106 |
+
"learning_rate": 8.600881834215168e-06,
|
| 107 |
+
"loss": 2.0722,
|
| 108 |
+
"step": 15876
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 15.0,
|
| 112 |
+
"grad_norm": 10.316903114318848,
|
| 113 |
+
"learning_rate": 8.500881834215169e-06,
|
| 114 |
+
"loss": 2.0726,
|
| 115 |
+
"step": 17010
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 16.0,
|
| 119 |
+
"grad_norm": 5.012073040008545,
|
| 120 |
+
"learning_rate": 8.400970017636686e-06,
|
| 121 |
+
"loss": 2.0657,
|
| 122 |
+
"step": 18144
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 17.0,
|
| 126 |
+
"grad_norm": 6.108023166656494,
|
| 127 |
+
"learning_rate": 8.301058201058203e-06,
|
| 128 |
+
"loss": 2.0583,
|
| 129 |
+
"step": 19278
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 18.0,
|
| 133 |
+
"grad_norm": 3.4143471717834473,
|
| 134 |
+
"learning_rate": 8.201058201058202e-06,
|
| 135 |
+
"loss": 2.038,
|
| 136 |
+
"step": 20412
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 19.0,
|
| 140 |
+
"grad_norm": 4.766075134277344,
|
| 141 |
+
"learning_rate": 8.101146384479719e-06,
|
| 142 |
+
"loss": 2.0292,
|
| 143 |
+
"step": 21546
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 20.0,
|
| 147 |
+
"grad_norm": 2.7671144008636475,
|
| 148 |
+
"learning_rate": 8.001146384479718e-06,
|
| 149 |
+
"loss": 2.0344,
|
| 150 |
+
"step": 22680
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 21.0,
|
| 154 |
+
"grad_norm": 3.447026491165161,
|
| 155 |
+
"learning_rate": 7.901234567901235e-06,
|
| 156 |
+
"loss": 2.0324,
|
| 157 |
+
"step": 23814
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 22.0,
|
| 161 |
+
"grad_norm": 3.207721471786499,
|
| 162 |
+
"learning_rate": 7.801234567901234e-06,
|
| 163 |
+
"loss": 2.0119,
|
| 164 |
+
"step": 24948
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 23.0,
|
| 168 |
+
"grad_norm": 2.789856433868408,
|
| 169 |
+
"learning_rate": 7.701234567901235e-06,
|
| 170 |
+
"loss": 2.0157,
|
| 171 |
+
"step": 26082
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 24.0,
|
| 175 |
+
"grad_norm": 2.2711899280548096,
|
| 176 |
+
"learning_rate": 7.601234567901235e-06,
|
| 177 |
+
"loss": 2.0048,
|
| 178 |
+
"step": 27216
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 25.0,
|
| 182 |
+
"grad_norm": 1.5879839658737183,
|
| 183 |
+
"learning_rate": 7.5013227513227514e-06,
|
| 184 |
+
"loss": 2.0111,
|
| 185 |
+
"step": 28350
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 26.0,
|
| 189 |
+
"grad_norm": 4.620934963226318,
|
| 190 |
+
"learning_rate": 7.401410934744268e-06,
|
| 191 |
+
"loss": 1.9837,
|
| 192 |
+
"step": 29484
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 27.0,
|
| 196 |
+
"grad_norm": 1.8832050561904907,
|
| 197 |
+
"learning_rate": 7.3014109347442685e-06,
|
| 198 |
+
"loss": 1.9818,
|
| 199 |
+
"step": 30618
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 28.0,
|
| 203 |
+
"grad_norm": 3.1029512882232666,
|
| 204 |
+
"learning_rate": 7.2014991181657855e-06,
|
| 205 |
+
"loss": 1.9868,
|
| 206 |
+
"step": 31752
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 29.0,
|
| 210 |
+
"grad_norm": 1.3787024021148682,
|
| 211 |
+
"learning_rate": 7.101499118165786e-06,
|
| 212 |
+
"loss": 1.988,
|
| 213 |
+
"step": 32886
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 30.0,
|
| 217 |
+
"grad_norm": 4.151069164276123,
|
| 218 |
+
"learning_rate": 7.001499118165786e-06,
|
| 219 |
+
"loss": 1.975,
|
| 220 |
+
"step": 34020
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 31.0,
|
| 224 |
+
"grad_norm": 3.1992034912109375,
|
| 225 |
+
"learning_rate": 6.901587301587303e-06,
|
| 226 |
+
"loss": 1.9814,
|
| 227 |
+
"step": 35154
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 32.0,
|
| 231 |
+
"grad_norm": 5.592874526977539,
|
| 232 |
+
"learning_rate": 6.801675485008819e-06,
|
| 233 |
+
"loss": 1.9739,
|
| 234 |
+
"step": 36288
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 33.0,
|
| 238 |
+
"grad_norm": 1.8030545711517334,
|
| 239 |
+
"learning_rate": 6.701675485008818e-06,
|
| 240 |
+
"loss": 1.9758,
|
| 241 |
+
"step": 37422
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 34.0,
|
| 245 |
+
"grad_norm": 3.9878218173980713,
|
| 246 |
+
"learning_rate": 6.601675485008819e-06,
|
| 247 |
+
"loss": 1.9726,
|
| 248 |
+
"step": 38556
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 35.0,
|
| 252 |
+
"grad_norm": 3.259308099746704,
|
| 253 |
+
"learning_rate": 6.501763668430335e-06,
|
| 254 |
+
"loss": 1.9649,
|
| 255 |
+
"step": 39690
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 36.0,
|
| 259 |
+
"grad_norm": 3.823164224624634,
|
| 260 |
+
"learning_rate": 6.401763668430335e-06,
|
| 261 |
+
"loss": 1.9547,
|
| 262 |
+
"step": 40824
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 37.0,
|
| 266 |
+
"grad_norm": 4.652733325958252,
|
| 267 |
+
"learning_rate": 6.301763668430335e-06,
|
| 268 |
+
"loss": 1.9673,
|
| 269 |
+
"step": 41958
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 38.0,
|
| 273 |
+
"grad_norm": 3.522451162338257,
|
| 274 |
+
"learning_rate": 6.201851851851852e-06,
|
| 275 |
+
"loss": 1.9544,
|
| 276 |
+
"step": 43092
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 39.0,
|
| 280 |
+
"grad_norm": 4.213411331176758,
|
| 281 |
+
"learning_rate": 6.1018518518518525e-06,
|
| 282 |
+
"loss": 1.9601,
|
| 283 |
+
"step": 44226
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 40.0,
|
| 287 |
+
"grad_norm": 5.826420783996582,
|
| 288 |
+
"learning_rate": 6.0019400352733694e-06,
|
| 289 |
+
"loss": 1.9553,
|
| 290 |
+
"step": 45360
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 41.0,
|
| 294 |
+
"grad_norm": 3.5787088871002197,
|
| 295 |
+
"learning_rate": 5.902028218694886e-06,
|
| 296 |
+
"loss": 1.9603,
|
| 297 |
+
"step": 46494
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 42.0,
|
| 301 |
+
"grad_norm": 2.380449056625366,
|
| 302 |
+
"learning_rate": 5.8020282186948865e-06,
|
| 303 |
+
"loss": 1.9551,
|
| 304 |
+
"step": 47628
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 43.0,
|
| 308 |
+
"grad_norm": 5.958215236663818,
|
| 309 |
+
"learning_rate": 5.702028218694886e-06,
|
| 310 |
+
"loss": 1.9515,
|
| 311 |
+
"step": 48762
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 44.0,
|
| 315 |
+
"grad_norm": 1.0750882625579834,
|
| 316 |
+
"learning_rate": 5.602116402116402e-06,
|
| 317 |
+
"loss": 1.9578,
|
| 318 |
+
"step": 49896
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 45.0,
|
| 322 |
+
"grad_norm": 2.135838031768799,
|
| 323 |
+
"learning_rate": 5.502116402116403e-06,
|
| 324 |
+
"loss": 1.9412,
|
| 325 |
+
"step": 51030
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 46.0,
|
| 329 |
+
"grad_norm": 2.9327504634857178,
|
| 330 |
+
"learning_rate": 5.402204585537919e-06,
|
| 331 |
+
"loss": 1.948,
|
| 332 |
+
"step": 52164
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 47.0,
|
| 336 |
+
"grad_norm": 5.212012767791748,
|
| 337 |
+
"learning_rate": 5.302204585537919e-06,
|
| 338 |
+
"loss": 1.9581,
|
| 339 |
+
"step": 53298
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 48.0,
|
| 343 |
+
"grad_norm": 3.5629754066467285,
|
| 344 |
+
"learning_rate": 5.202292768959436e-06,
|
| 345 |
+
"loss": 1.9394,
|
| 346 |
+
"step": 54432
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 49.0,
|
| 350 |
+
"grad_norm": 2.700446844100952,
|
| 351 |
+
"learning_rate": 5.102292768959436e-06,
|
| 352 |
+
"loss": 1.943,
|
| 353 |
+
"step": 55566
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 50.0,
|
| 357 |
+
"grad_norm": 5.429810523986816,
|
| 358 |
+
"learning_rate": 5.002380952380953e-06,
|
| 359 |
+
"loss": 1.9499,
|
| 360 |
+
"step": 56700
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 51.0,
|
| 364 |
+
"grad_norm": 2.2767255306243896,
|
| 365 |
+
"learning_rate": 4.902469135802469e-06,
|
| 366 |
+
"loss": 1.933,
|
| 367 |
+
"step": 57834
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 52.0,
|
| 371 |
+
"grad_norm": 4.359825611114502,
|
| 372 |
+
"learning_rate": 4.802469135802469e-06,
|
| 373 |
+
"loss": 1.9319,
|
| 374 |
+
"step": 58968
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 53.0,
|
| 378 |
+
"grad_norm": 3.3374125957489014,
|
| 379 |
+
"learning_rate": 4.702557319223986e-06,
|
| 380 |
+
"loss": 1.9303,
|
| 381 |
+
"step": 60102
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 54.0,
|
| 385 |
+
"grad_norm": 1.666292428970337,
|
| 386 |
+
"learning_rate": 4.6025573192239865e-06,
|
| 387 |
+
"loss": 1.9391,
|
| 388 |
+
"step": 61236
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 55.0,
|
| 392 |
+
"grad_norm": 2.920473575592041,
|
| 393 |
+
"learning_rate": 4.502645502645503e-06,
|
| 394 |
+
"loss": 1.9339,
|
| 395 |
+
"step": 62370
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 56.0,
|
| 399 |
+
"grad_norm": 3.5226597785949707,
|
| 400 |
+
"learning_rate": 4.402645502645503e-06,
|
| 401 |
+
"loss": 1.924,
|
| 402 |
+
"step": 63504
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 57.0,
|
| 406 |
+
"grad_norm": 1.056983232498169,
|
| 407 |
+
"learning_rate": 4.30273368606702e-06,
|
| 408 |
+
"loss": 1.939,
|
| 409 |
+
"step": 64638
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 58.0,
|
| 413 |
+
"grad_norm": 1.7839380502700806,
|
| 414 |
+
"learning_rate": 4.20273368606702e-06,
|
| 415 |
+
"loss": 1.9302,
|
| 416 |
+
"step": 65772
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 59.0,
|
| 420 |
+
"grad_norm": 2.453362464904785,
|
| 421 |
+
"learning_rate": 4.102821869488536e-06,
|
| 422 |
+
"loss": 1.932,
|
| 423 |
+
"step": 66906
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 60.0,
|
| 427 |
+
"grad_norm": 1.79325532913208,
|
| 428 |
+
"learning_rate": 4.002910052910054e-06,
|
| 429 |
+
"loss": 1.9347,
|
| 430 |
+
"step": 68040
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 61.0,
|
| 434 |
+
"grad_norm": 4.606123924255371,
|
| 435 |
+
"learning_rate": 3.902910052910053e-06,
|
| 436 |
+
"loss": 1.9163,
|
| 437 |
+
"step": 69174
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 62.0,
|
| 441 |
+
"grad_norm": 3.7879464626312256,
|
| 442 |
+
"learning_rate": 3.80299823633157e-06,
|
| 443 |
+
"loss": 1.9314,
|
| 444 |
+
"step": 70308
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 63.0,
|
| 448 |
+
"grad_norm": 3.875440835952759,
|
| 449 |
+
"learning_rate": 3.7029982363315697e-06,
|
| 450 |
+
"loss": 1.9364,
|
| 451 |
+
"step": 71442
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 64.0,
|
| 455 |
+
"grad_norm": 3.8616816997528076,
|
| 456 |
+
"learning_rate": 3.6030864197530867e-06,
|
| 457 |
+
"loss": 1.9183,
|
| 458 |
+
"step": 72576
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 65.0,
|
| 462 |
+
"grad_norm": 3.232619285583496,
|
| 463 |
+
"learning_rate": 3.5030864197530868e-06,
|
| 464 |
+
"loss": 1.9164,
|
| 465 |
+
"step": 73710
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 66.0,
|
| 469 |
+
"grad_norm": 3.7224302291870117,
|
| 470 |
+
"learning_rate": 3.4031746031746033e-06,
|
| 471 |
+
"loss": 1.9225,
|
| 472 |
+
"step": 74844
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 67.0,
|
| 476 |
+
"grad_norm": 2.774463653564453,
|
| 477 |
+
"learning_rate": 3.303174603174604e-06,
|
| 478 |
+
"loss": 1.9328,
|
| 479 |
+
"step": 75978
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 68.0,
|
| 483 |
+
"grad_norm": 1.1616075038909912,
|
| 484 |
+
"learning_rate": 3.2032627865961204e-06,
|
| 485 |
+
"loss": 1.914,
|
| 486 |
+
"step": 77112
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 69.0,
|
| 490 |
+
"grad_norm": 1.613097071647644,
|
| 491 |
+
"learning_rate": 3.103350970017637e-06,
|
| 492 |
+
"loss": 1.9147,
|
| 493 |
+
"step": 78246
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 70.0,
|
| 497 |
+
"grad_norm": 4.327018737792969,
|
| 498 |
+
"learning_rate": 3.003350970017637e-06,
|
| 499 |
+
"loss": 1.924,
|
| 500 |
+
"step": 79380
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 71.0,
|
| 504 |
+
"grad_norm": 1.2719597816467285,
|
| 505 |
+
"learning_rate": 2.9034391534391536e-06,
|
| 506 |
+
"loss": 1.938,
|
| 507 |
+
"step": 80514
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 72.0,
|
| 511 |
+
"grad_norm": 2.238760232925415,
|
| 512 |
+
"learning_rate": 2.8034391534391537e-06,
|
| 513 |
+
"loss": 1.9153,
|
| 514 |
+
"step": 81648
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 73.0,
|
| 518 |
+
"grad_norm": 1.070393443107605,
|
| 519 |
+
"learning_rate": 2.7035273368606703e-06,
|
| 520 |
+
"loss": 1.9171,
|
| 521 |
+
"step": 82782
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 74.0,
|
| 525 |
+
"grad_norm": 3.3792688846588135,
|
| 526 |
+
"learning_rate": 2.603527336860671e-06,
|
| 527 |
+
"loss": 1.9204,
|
| 528 |
+
"step": 83916
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 75.0,
|
| 532 |
+
"grad_norm": 1.406660795211792,
|
| 533 |
+
"learning_rate": 2.503615520282187e-06,
|
| 534 |
+
"loss": 1.9139,
|
| 535 |
+
"step": 85050
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 76.0,
|
| 539 |
+
"grad_norm": 1.7133662700653076,
|
| 540 |
+
"learning_rate": 2.403615520282187e-06,
|
| 541 |
+
"loss": 1.9238,
|
| 542 |
+
"step": 86184
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 77.0,
|
| 546 |
+
"grad_norm": 1.5862336158752441,
|
| 547 |
+
"learning_rate": 2.303703703703704e-06,
|
| 548 |
+
"loss": 1.9193,
|
| 549 |
+
"step": 87318
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 78.0,
|
| 553 |
+
"grad_norm": 2.752800226211548,
|
| 554 |
+
"learning_rate": 2.2037918871252206e-06,
|
| 555 |
+
"loss": 1.9147,
|
| 556 |
+
"step": 88452
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 79.0,
|
| 560 |
+
"grad_norm": 1.8123775720596313,
|
| 561 |
+
"learning_rate": 2.1037918871252207e-06,
|
| 562 |
+
"loss": 1.9007,
|
| 563 |
+
"step": 89586
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 80.0,
|
| 567 |
+
"grad_norm": 3.6397151947021484,
|
| 568 |
+
"learning_rate": 2.0038800705467372e-06,
|
| 569 |
+
"loss": 1.9123,
|
| 570 |
+
"step": 90720
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 81.0,
|
| 574 |
+
"grad_norm": 2.8387656211853027,
|
| 575 |
+
"learning_rate": 1.9038800705467373e-06,
|
| 576 |
+
"loss": 1.9165,
|
| 577 |
+
"step": 91854
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 82.0,
|
| 581 |
+
"grad_norm": 0.9121940732002258,
|
| 582 |
+
"learning_rate": 1.803968253968254e-06,
|
| 583 |
+
"loss": 1.9179,
|
| 584 |
+
"step": 92988
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 83.0,
|
| 588 |
+
"grad_norm": 3.447791576385498,
|
| 589 |
+
"learning_rate": 1.7039682539682542e-06,
|
| 590 |
+
"loss": 1.9049,
|
| 591 |
+
"step": 94122
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 84.0,
|
| 595 |
+
"grad_norm": 2.4387011528015137,
|
| 596 |
+
"learning_rate": 1.6040564373897708e-06,
|
| 597 |
+
"loss": 1.9186,
|
| 598 |
+
"step": 95256
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 85.0,
|
| 602 |
+
"grad_norm": 4.287496566772461,
|
| 603 |
+
"learning_rate": 1.5040564373897709e-06,
|
| 604 |
+
"loss": 1.9137,
|
| 605 |
+
"step": 96390
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 86.0,
|
| 609 |
+
"grad_norm": 1.9334492683410645,
|
| 610 |
+
"learning_rate": 1.4041446208112876e-06,
|
| 611 |
+
"loss": 1.9245,
|
| 612 |
+
"step": 97524
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 87.0,
|
| 616 |
+
"grad_norm": 3.3120267391204834,
|
| 617 |
+
"learning_rate": 1.3042328042328042e-06,
|
| 618 |
+
"loss": 1.9022,
|
| 619 |
+
"step": 98658
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 88.0,
|
| 623 |
+
"grad_norm": 3.476644992828369,
|
| 624 |
+
"learning_rate": 1.2042328042328043e-06,
|
| 625 |
+
"loss": 1.9157,
|
| 626 |
+
"step": 99792
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 89.0,
|
| 630 |
+
"grad_norm": 1.7216417789459229,
|
| 631 |
+
"learning_rate": 1.104320987654321e-06,
|
| 632 |
+
"loss": 1.9212,
|
| 633 |
+
"step": 100926
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 90.0,
|
| 637 |
+
"grad_norm": 1.2965023517608643,
|
| 638 |
+
"learning_rate": 1.0043209876543212e-06,
|
| 639 |
+
"loss": 1.9098,
|
| 640 |
+
"step": 102060
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 91.0,
|
| 644 |
+
"grad_norm": 2.355569362640381,
|
| 645 |
+
"learning_rate": 9.044091710758378e-07,
|
| 646 |
+
"loss": 1.9128,
|
| 647 |
+
"step": 103194
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 92.0,
|
| 651 |
+
"grad_norm": 1.394773006439209,
|
| 652 |
+
"learning_rate": 8.044091710758378e-07,
|
| 653 |
+
"loss": 1.9084,
|
| 654 |
+
"step": 104328
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 93.0,
|
| 658 |
+
"grad_norm": 2.99224591255188,
|
| 659 |
+
"learning_rate": 7.044973544973546e-07,
|
| 660 |
+
"loss": 1.902,
|
| 661 |
+
"step": 105462
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 94.0,
|
| 665 |
+
"grad_norm": 1.2547388076782227,
|
| 666 |
+
"learning_rate": 6.044973544973545e-07,
|
| 667 |
+
"loss": 1.9185,
|
| 668 |
+
"step": 106596
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 95.0,
|
| 672 |
+
"grad_norm": 0.7210695147514343,
|
| 673 |
+
"learning_rate": 5.045855379188714e-07,
|
| 674 |
+
"loss": 1.9131,
|
| 675 |
+
"step": 107730
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 96.0,
|
| 679 |
+
"grad_norm": 0.7888159155845642,
|
| 680 |
+
"learning_rate": 4.0467372134038806e-07,
|
| 681 |
+
"loss": 1.9087,
|
| 682 |
+
"step": 108864
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 97.0,
|
| 686 |
+
"grad_norm": 0.9536716341972351,
|
| 687 |
+
"learning_rate": 3.04673721340388e-07,
|
| 688 |
+
"loss": 1.9029,
|
| 689 |
+
"step": 109998
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 98.0,
|
| 693 |
+
"grad_norm": 2.2886030673980713,
|
| 694 |
+
"learning_rate": 2.047619047619048e-07,
|
| 695 |
+
"loss": 1.9094,
|
| 696 |
+
"step": 111132
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 99.0,
|
| 700 |
+
"grad_norm": 2.0867316722869873,
|
| 701 |
+
"learning_rate": 1.0476190476190476e-07,
|
| 702 |
+
"loss": 1.9115,
|
| 703 |
+
"step": 112266
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 100.0,
|
| 707 |
+
"grad_norm": 0.59788578748703,
|
| 708 |
+
"learning_rate": 4.8500881834215175e-09,
|
| 709 |
+
"loss": 1.8996,
|
| 710 |
+
"step": 113400
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 100.0,
|
| 714 |
+
"step": 113400,
|
| 715 |
+
"total_flos": 3.571154905966387e+18,
|
| 716 |
+
"train_loss": 1.9808363689442792,
|
| 717 |
+
"train_runtime": 84553.2908,
|
| 718 |
+
"train_samples_per_second": 160.897,
|
| 719 |
+
"train_steps_per_second": 1.341
|
| 720 |
+
}
|
| 721 |
+
],
|
| 722 |
+
"logging_steps": 1134,
|
| 723 |
+
"max_steps": 113400,
|
| 724 |
+
"num_input_tokens_seen": 0,
|
| 725 |
+
"num_train_epochs": 100,
|
| 726 |
+
"save_steps": 530,
|
| 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": 3.571154905966387e+18,
|
| 740 |
+
"train_batch_size": 60,
|
| 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
|
|
|