Public Release
Browse files- all_results.json +15 -0
- config.json +26 -0
- eval_results.json +10 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +14 -0
- train_results.json +8 -0
- trainer_state.json +775 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 74.94,
|
| 3 |
+
"eval_accuracy": 0.8010044122316655,
|
| 4 |
+
"eval_loss": 0.82177734375,
|
| 5 |
+
"eval_runtime": 370.935,
|
| 6 |
+
"eval_samples": 143987,
|
| 7 |
+
"eval_samples_per_second": 388.173,
|
| 8 |
+
"eval_steps_per_second": 1.518,
|
| 9 |
+
"perplexity": 2.274538884552246,
|
| 10 |
+
"train_loss": 0.06944640869140625,
|
| 11 |
+
"train_runtime": 3177.0339,
|
| 12 |
+
"train_samples": 3414625,
|
| 13 |
+
"train_samples_per_second": 80578.303,
|
| 14 |
+
"train_steps_per_second": 39.345
|
| 15 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/grand/projects/SolarWindowsADSP/pankaj/polaris_outputs/BASE-TOKENIZER-cased-from-bert-base/checkpoint-115000",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"gradient_checkpointing": false,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 512,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"torch_dtype": "float16",
|
| 22 |
+
"transformers_version": "4.25.1",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 28996
|
| 26 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 74.94,
|
| 3 |
+
"eval_accuracy": 0.8010044122316655,
|
| 4 |
+
"eval_loss": 0.82177734375,
|
| 5 |
+
"eval_runtime": 370.935,
|
| 6 |
+
"eval_samples": 143987,
|
| 7 |
+
"eval_samples_per_second": 388.173,
|
| 8 |
+
"eval_steps_per_second": 1.518,
|
| 9 |
+
"perplexity": 2.274538884552246
|
| 10 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:63a01b8fae3b94545df3db396db58e799f04b2830603958a662cb7e5a517ba53
|
| 3 |
+
size 216714531
|
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 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"do_lower_case": false,
|
| 4 |
+
"mask_token": "[MASK]",
|
| 5 |
+
"model_max_length": 512,
|
| 6 |
+
"name_or_path": "/grand/projects/SolarWindowsADSP/pankaj/polaris_outputs/BASE-TOKENIZER-cased-from-bert-base/checkpoint-115000",
|
| 7 |
+
"pad_token": "[PAD]",
|
| 8 |
+
"sep_token": "[SEP]",
|
| 9 |
+
"special_tokens_map_file": null,
|
| 10 |
+
"strip_accents": null,
|
| 11 |
+
"tokenize_chinese_chars": true,
|
| 12 |
+
"tokenizer_class": "BertTokenizer",
|
| 13 |
+
"unk_token": "[UNK]"
|
| 14 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 74.94,
|
| 3 |
+
"train_loss": 0.06944640869140625,
|
| 4 |
+
"train_runtime": 3177.0339,
|
| 5 |
+
"train_samples": 3414625,
|
| 6 |
+
"train_samples_per_second": 80578.303,
|
| 7 |
+
"train_steps_per_second": 39.345
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,775 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 74.9400479616307,
|
| 5 |
+
"global_step": 125000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.6,
|
| 12 |
+
"learning_rate": 7.832e-06,
|
| 13 |
+
"loss": 3.1441,
|
| 14 |
+
"step": 1000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.2,
|
| 18 |
+
"learning_rate": 1.5832e-05,
|
| 19 |
+
"loss": 1.8595,
|
| 20 |
+
"step": 2000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.8,
|
| 24 |
+
"learning_rate": 2.3832e-05,
|
| 25 |
+
"loss": 1.5378,
|
| 26 |
+
"step": 3000
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.4,
|
| 30 |
+
"learning_rate": 3.1832e-05,
|
| 31 |
+
"loss": 1.43,
|
| 32 |
+
"step": 4000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 3.0,
|
| 36 |
+
"learning_rate": 3.9816e-05,
|
| 37 |
+
"loss": 1.3543,
|
| 38 |
+
"step": 5000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 3.6,
|
| 42 |
+
"learning_rate": 4.7816000000000004e-05,
|
| 43 |
+
"loss": 1.2961,
|
| 44 |
+
"step": 6000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 4.2,
|
| 48 |
+
"learning_rate": 5.5808000000000005e-05,
|
| 49 |
+
"loss": 1.2511,
|
| 50 |
+
"step": 7000
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.8,
|
| 54 |
+
"learning_rate": 6.380800000000001e-05,
|
| 55 |
+
"loss": 1.2147,
|
| 56 |
+
"step": 8000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 5.4,
|
| 60 |
+
"learning_rate": 7.1808e-05,
|
| 61 |
+
"loss": 1.1854,
|
| 62 |
+
"step": 9000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 6.0,
|
| 66 |
+
"learning_rate": 7.980000000000002e-05,
|
| 67 |
+
"loss": 1.1596,
|
| 68 |
+
"step": 10000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 6.59,
|
| 72 |
+
"learning_rate": 7.93224347826087e-05,
|
| 73 |
+
"loss": 1.1377,
|
| 74 |
+
"step": 11000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 7.19,
|
| 78 |
+
"learning_rate": 7.862747826086958e-05,
|
| 79 |
+
"loss": 1.1184,
|
| 80 |
+
"step": 12000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 7.79,
|
| 84 |
+
"learning_rate": 7.793252173913044e-05,
|
| 85 |
+
"loss": 1.102,
|
| 86 |
+
"step": 13000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 8.39,
|
| 90 |
+
"learning_rate": 7.72368695652174e-05,
|
| 91 |
+
"loss": 1.0876,
|
| 92 |
+
"step": 14000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 8.99,
|
| 96 |
+
"learning_rate": 7.654191304347826e-05,
|
| 97 |
+
"loss": 1.0765,
|
| 98 |
+
"step": 15000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 9.59,
|
| 102 |
+
"learning_rate": 7.584695652173914e-05,
|
| 103 |
+
"loss": 1.0655,
|
| 104 |
+
"step": 16000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 10.19,
|
| 108 |
+
"learning_rate": 7.5152e-05,
|
| 109 |
+
"loss": 1.0555,
|
| 110 |
+
"step": 17000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 10.79,
|
| 114 |
+
"learning_rate": 7.445773913043479e-05,
|
| 115 |
+
"loss": 1.0478,
|
| 116 |
+
"step": 18000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 11.39,
|
| 120 |
+
"learning_rate": 7.376208695652175e-05,
|
| 121 |
+
"loss": 1.0403,
|
| 122 |
+
"step": 19000
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 11.99,
|
| 126 |
+
"learning_rate": 7.306713043478261e-05,
|
| 127 |
+
"loss": 1.0334,
|
| 128 |
+
"step": 20000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 12.59,
|
| 132 |
+
"learning_rate": 7.237217391304349e-05,
|
| 133 |
+
"loss": 1.026,
|
| 134 |
+
"step": 21000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 13.19,
|
| 138 |
+
"learning_rate": 7.167791304347826e-05,
|
| 139 |
+
"loss": 1.0195,
|
| 140 |
+
"step": 22000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 13.79,
|
| 144 |
+
"learning_rate": 7.098226086956523e-05,
|
| 145 |
+
"loss": 1.0144,
|
| 146 |
+
"step": 23000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 14.39,
|
| 150 |
+
"learning_rate": 7.028730434782609e-05,
|
| 151 |
+
"loss": 1.0086,
|
| 152 |
+
"step": 24000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 14.99,
|
| 156 |
+
"learning_rate": 6.959234782608696e-05,
|
| 157 |
+
"loss": 1.0045,
|
| 158 |
+
"step": 25000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 15.59,
|
| 162 |
+
"learning_rate": 8.000000000000001e-06,
|
| 163 |
+
"loss": 0.9939,
|
| 164 |
+
"step": 26000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 16.19,
|
| 168 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 169 |
+
"loss": 0.989,
|
| 170 |
+
"step": 27000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 16.79,
|
| 174 |
+
"learning_rate": 2.4e-05,
|
| 175 |
+
"loss": 0.9883,
|
| 176 |
+
"step": 28000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 17.39,
|
| 180 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 181 |
+
"loss": 0.9875,
|
| 182 |
+
"step": 29000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 17.99,
|
| 186 |
+
"learning_rate": 4e-05,
|
| 187 |
+
"loss": 0.9861,
|
| 188 |
+
"step": 30000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 18.59,
|
| 192 |
+
"learning_rate": 4.8e-05,
|
| 193 |
+
"loss": 0.9855,
|
| 194 |
+
"step": 31000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 19.18,
|
| 198 |
+
"learning_rate": 5.6e-05,
|
| 199 |
+
"loss": 0.9845,
|
| 200 |
+
"step": 32000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 19.78,
|
| 204 |
+
"learning_rate": 6.400000000000001e-05,
|
| 205 |
+
"loss": 0.9839,
|
| 206 |
+
"step": 33000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 20.38,
|
| 210 |
+
"learning_rate": 7.2e-05,
|
| 211 |
+
"loss": 0.9832,
|
| 212 |
+
"step": 34000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 20.98,
|
| 216 |
+
"learning_rate": 8e-05,
|
| 217 |
+
"loss": 0.982,
|
| 218 |
+
"step": 35000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 21.58,
|
| 222 |
+
"learning_rate": 7.930434782608697e-05,
|
| 223 |
+
"loss": 0.9802,
|
| 224 |
+
"step": 36000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 22.18,
|
| 228 |
+
"learning_rate": 7.860869565217392e-05,
|
| 229 |
+
"loss": 0.9762,
|
| 230 |
+
"step": 37000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 22.78,
|
| 234 |
+
"learning_rate": 7.791304347826088e-05,
|
| 235 |
+
"loss": 0.9723,
|
| 236 |
+
"step": 38000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 23.38,
|
| 240 |
+
"learning_rate": 7.721739130434783e-05,
|
| 241 |
+
"loss": 0.9698,
|
| 242 |
+
"step": 39000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 23.98,
|
| 246 |
+
"learning_rate": 7.652173913043479e-05,
|
| 247 |
+
"loss": 0.9667,
|
| 248 |
+
"step": 40000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 24.58,
|
| 252 |
+
"learning_rate": 7.582608695652174e-05,
|
| 253 |
+
"loss": 0.9637,
|
| 254 |
+
"step": 41000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 25.18,
|
| 258 |
+
"learning_rate": 7.51304347826087e-05,
|
| 259 |
+
"loss": 0.9614,
|
| 260 |
+
"step": 42000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 25.78,
|
| 264 |
+
"learning_rate": 7.443478260869565e-05,
|
| 265 |
+
"loss": 0.9578,
|
| 266 |
+
"step": 43000
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 26.38,
|
| 270 |
+
"learning_rate": 7.373913043478261e-05,
|
| 271 |
+
"loss": 0.9553,
|
| 272 |
+
"step": 44000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 26.98,
|
| 276 |
+
"learning_rate": 7.304347826086957e-05,
|
| 277 |
+
"loss": 0.9532,
|
| 278 |
+
"step": 45000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 27.58,
|
| 282 |
+
"learning_rate": 7.856e-06,
|
| 283 |
+
"loss": 0.9441,
|
| 284 |
+
"step": 46000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 28.18,
|
| 288 |
+
"learning_rate": 1.5856e-05,
|
| 289 |
+
"loss": 0.9399,
|
| 290 |
+
"step": 47000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 28.78,
|
| 294 |
+
"learning_rate": 2.3840000000000002e-05,
|
| 295 |
+
"loss": 0.9378,
|
| 296 |
+
"step": 48000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 29.38,
|
| 300 |
+
"learning_rate": 3.184000000000001e-05,
|
| 301 |
+
"loss": 0.9373,
|
| 302 |
+
"step": 49000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 29.98,
|
| 306 |
+
"learning_rate": 3.9824000000000004e-05,
|
| 307 |
+
"loss": 0.9379,
|
| 308 |
+
"step": 50000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 30.58,
|
| 312 |
+
"learning_rate": 4.7824e-05,
|
| 313 |
+
"loss": 0.9376,
|
| 314 |
+
"step": 51000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 31.18,
|
| 318 |
+
"learning_rate": 5.5816e-05,
|
| 319 |
+
"loss": 0.9384,
|
| 320 |
+
"step": 52000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 31.77,
|
| 324 |
+
"learning_rate": 6.380800000000001e-05,
|
| 325 |
+
"loss": 0.9393,
|
| 326 |
+
"step": 53000
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 32.37,
|
| 330 |
+
"learning_rate": 7.1808e-05,
|
| 331 |
+
"loss": 0.9401,
|
| 332 |
+
"step": 54000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 32.97,
|
| 336 |
+
"learning_rate": 7.9792e-05,
|
| 337 |
+
"loss": 0.9397,
|
| 338 |
+
"step": 55000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 33.57,
|
| 342 |
+
"learning_rate": 7.932313043478262e-05,
|
| 343 |
+
"loss": 0.9393,
|
| 344 |
+
"step": 56000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 34.17,
|
| 348 |
+
"learning_rate": 7.862747826086958e-05,
|
| 349 |
+
"loss": 0.9383,
|
| 350 |
+
"step": 57000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 34.77,
|
| 354 |
+
"learning_rate": 7.793182608695653e-05,
|
| 355 |
+
"loss": 0.9362,
|
| 356 |
+
"step": 58000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 35.37,
|
| 360 |
+
"learning_rate": 7.72368695652174e-05,
|
| 361 |
+
"loss": 0.9342,
|
| 362 |
+
"step": 59000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 35.97,
|
| 366 |
+
"learning_rate": 7.654191304347826e-05,
|
| 367 |
+
"loss": 0.933,
|
| 368 |
+
"step": 60000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 36.57,
|
| 372 |
+
"learning_rate": 7.584626086956523e-05,
|
| 373 |
+
"loss": 0.9295,
|
| 374 |
+
"step": 61000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 37.17,
|
| 378 |
+
"learning_rate": 7.51513043478261e-05,
|
| 379 |
+
"loss": 0.9283,
|
| 380 |
+
"step": 62000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 37.77,
|
| 384 |
+
"learning_rate": 7.445565217391305e-05,
|
| 385 |
+
"loss": 0.9268,
|
| 386 |
+
"step": 63000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 38.37,
|
| 390 |
+
"learning_rate": 7.376069565217393e-05,
|
| 391 |
+
"loss": 0.9252,
|
| 392 |
+
"step": 64000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 38.97,
|
| 396 |
+
"learning_rate": 7.306573913043479e-05,
|
| 397 |
+
"loss": 0.923,
|
| 398 |
+
"step": 65000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 39.57,
|
| 402 |
+
"learning_rate": 7.237147826086958e-05,
|
| 403 |
+
"loss": 0.922,
|
| 404 |
+
"step": 66000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 40.17,
|
| 408 |
+
"learning_rate": 7.167652173913044e-05,
|
| 409 |
+
"loss": 0.9201,
|
| 410 |
+
"step": 67000
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 40.77,
|
| 414 |
+
"learning_rate": 7.09808695652174e-05,
|
| 415 |
+
"loss": 0.9182,
|
| 416 |
+
"step": 68000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 41.37,
|
| 420 |
+
"learning_rate": 7.028591304347828e-05,
|
| 421 |
+
"loss": 0.9166,
|
| 422 |
+
"step": 69000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 41.97,
|
| 426 |
+
"learning_rate": 6.959095652173914e-05,
|
| 427 |
+
"loss": 0.9159,
|
| 428 |
+
"step": 70000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 42.57,
|
| 432 |
+
"learning_rate": 6.8896e-05,
|
| 433 |
+
"loss": 0.9135,
|
| 434 |
+
"step": 71000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 43.17,
|
| 438 |
+
"learning_rate": 6.820104347826087e-05,
|
| 439 |
+
"loss": 0.9131,
|
| 440 |
+
"step": 72000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 43.76,
|
| 444 |
+
"learning_rate": 6.750539130434784e-05,
|
| 445 |
+
"loss": 0.9104,
|
| 446 |
+
"step": 73000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 44.36,
|
| 450 |
+
"learning_rate": 6.680973913043478e-05,
|
| 451 |
+
"loss": 0.91,
|
| 452 |
+
"step": 74000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 44.96,
|
| 456 |
+
"learning_rate": 6.611547826086957e-05,
|
| 457 |
+
"loss": 0.9084,
|
| 458 |
+
"step": 75000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 45.56,
|
| 462 |
+
"learning_rate": 6.541982608695652e-05,
|
| 463 |
+
"loss": 0.9072,
|
| 464 |
+
"step": 76000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 46.16,
|
| 468 |
+
"learning_rate": 6.472556521739131e-05,
|
| 469 |
+
"loss": 0.9065,
|
| 470 |
+
"step": 77000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 46.76,
|
| 474 |
+
"learning_rate": 6.402991304347827e-05,
|
| 475 |
+
"loss": 0.9041,
|
| 476 |
+
"step": 78000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 47.36,
|
| 480 |
+
"learning_rate": 6.333426086956522e-05,
|
| 481 |
+
"loss": 0.9029,
|
| 482 |
+
"step": 79000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 47.96,
|
| 486 |
+
"learning_rate": 6.264000000000001e-05,
|
| 487 |
+
"loss": 0.9022,
|
| 488 |
+
"step": 80000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 48.56,
|
| 492 |
+
"learning_rate": 6.194434782608696e-05,
|
| 493 |
+
"loss": 0.9007,
|
| 494 |
+
"step": 81000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 49.16,
|
| 498 |
+
"learning_rate": 6.124939130434783e-05,
|
| 499 |
+
"loss": 0.8998,
|
| 500 |
+
"step": 82000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 49.76,
|
| 504 |
+
"learning_rate": 6.05544347826087e-05,
|
| 505 |
+
"loss": 0.8986,
|
| 506 |
+
"step": 83000
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 50.36,
|
| 510 |
+
"learning_rate": 5.985878260869566e-05,
|
| 511 |
+
"loss": 0.8966,
|
| 512 |
+
"step": 84000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 50.96,
|
| 516 |
+
"learning_rate": 5.916382608695653e-05,
|
| 517 |
+
"loss": 0.8967,
|
| 518 |
+
"step": 85000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 51.56,
|
| 522 |
+
"learning_rate": 5.8468869565217395e-05,
|
| 523 |
+
"loss": 0.8952,
|
| 524 |
+
"step": 86000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 52.16,
|
| 528 |
+
"learning_rate": 5.777321739130435e-05,
|
| 529 |
+
"loss": 0.8937,
|
| 530 |
+
"step": 87000
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 52.76,
|
| 534 |
+
"learning_rate": 5.7078260869565226e-05,
|
| 535 |
+
"loss": 0.8934,
|
| 536 |
+
"step": 88000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 53.36,
|
| 540 |
+
"learning_rate": 5.6382608695652174e-05,
|
| 541 |
+
"loss": 0.8928,
|
| 542 |
+
"step": 89000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 53.96,
|
| 546 |
+
"learning_rate": 5.568765217391304e-05,
|
| 547 |
+
"loss": 0.8909,
|
| 548 |
+
"step": 90000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 54.56,
|
| 552 |
+
"learning_rate": 5.499269565217392e-05,
|
| 553 |
+
"loss": 0.89,
|
| 554 |
+
"step": 91000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 55.16,
|
| 558 |
+
"learning_rate": 5.429773913043479e-05,
|
| 559 |
+
"loss": 0.8889,
|
| 560 |
+
"step": 92000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 55.76,
|
| 564 |
+
"learning_rate": 5.360208695652175e-05,
|
| 565 |
+
"loss": 0.8883,
|
| 566 |
+
"step": 93000
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 56.35,
|
| 570 |
+
"learning_rate": 5.2907826086956524e-05,
|
| 571 |
+
"loss": 0.8873,
|
| 572 |
+
"step": 94000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 56.95,
|
| 576 |
+
"learning_rate": 5.22128695652174e-05,
|
| 577 |
+
"loss": 0.8862,
|
| 578 |
+
"step": 95000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 57.55,
|
| 582 |
+
"learning_rate": 5.151721739130435e-05,
|
| 583 |
+
"loss": 0.8857,
|
| 584 |
+
"step": 96000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 58.15,
|
| 588 |
+
"learning_rate": 5.082156521739131e-05,
|
| 589 |
+
"loss": 0.8849,
|
| 590 |
+
"step": 97000
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 58.75,
|
| 594 |
+
"learning_rate": 5.012660869565218e-05,
|
| 595 |
+
"loss": 0.8829,
|
| 596 |
+
"step": 98000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 59.35,
|
| 600 |
+
"learning_rate": 4.943234782608696e-05,
|
| 601 |
+
"loss": 0.8836,
|
| 602 |
+
"step": 99000
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 59.95,
|
| 606 |
+
"learning_rate": 4.873669565217392e-05,
|
| 607 |
+
"loss": 0.8818,
|
| 608 |
+
"step": 100000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 60.55,
|
| 612 |
+
"learning_rate": 4.804173913043479e-05,
|
| 613 |
+
"loss": 0.8811,
|
| 614 |
+
"step": 101000
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 61.15,
|
| 618 |
+
"learning_rate": 4.734608695652174e-05,
|
| 619 |
+
"loss": 0.8805,
|
| 620 |
+
"step": 102000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 61.75,
|
| 624 |
+
"learning_rate": 4.6651130434782615e-05,
|
| 625 |
+
"loss": 0.8797,
|
| 626 |
+
"step": 103000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 62.35,
|
| 630 |
+
"learning_rate": 4.595547826086957e-05,
|
| 631 |
+
"loss": 0.8791,
|
| 632 |
+
"step": 104000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 62.95,
|
| 636 |
+
"learning_rate": 4.526121739130435e-05,
|
| 637 |
+
"loss": 0.8782,
|
| 638 |
+
"step": 105000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 63.55,
|
| 642 |
+
"learning_rate": 4.456626086956522e-05,
|
| 643 |
+
"loss": 0.8767,
|
| 644 |
+
"step": 106000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 64.15,
|
| 648 |
+
"learning_rate": 4.3870608695652176e-05,
|
| 649 |
+
"loss": 0.8769,
|
| 650 |
+
"step": 107000
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 64.75,
|
| 654 |
+
"learning_rate": 4.317495652173914e-05,
|
| 655 |
+
"loss": 0.8754,
|
| 656 |
+
"step": 108000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 65.35,
|
| 660 |
+
"learning_rate": 4.248000000000001e-05,
|
| 661 |
+
"loss": 0.8746,
|
| 662 |
+
"step": 109000
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 65.95,
|
| 666 |
+
"learning_rate": 4.1785043478260875e-05,
|
| 667 |
+
"loss": 0.8752,
|
| 668 |
+
"step": 110000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 66.55,
|
| 672 |
+
"learning_rate": 4.108939130434783e-05,
|
| 673 |
+
"loss": 0.8735,
|
| 674 |
+
"step": 111000
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 67.15,
|
| 678 |
+
"learning_rate": 4.0395130434782606e-05,
|
| 679 |
+
"loss": 0.8733,
|
| 680 |
+
"step": 112000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 67.75,
|
| 684 |
+
"learning_rate": 3.970017391304348e-05,
|
| 685 |
+
"loss": 0.8727,
|
| 686 |
+
"step": 113000
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 68.35,
|
| 690 |
+
"learning_rate": 3.900521739130435e-05,
|
| 691 |
+
"loss": 0.8715,
|
| 692 |
+
"step": 114000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 68.94,
|
| 696 |
+
"learning_rate": 3.830956521739131e-05,
|
| 697 |
+
"loss": 0.8721,
|
| 698 |
+
"step": 115000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 69.54,
|
| 702 |
+
"learning_rate": 3.7614608695652174e-05,
|
| 703 |
+
"loss": 0.8712,
|
| 704 |
+
"step": 116000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 70.14,
|
| 708 |
+
"learning_rate": 3.691965217391304e-05,
|
| 709 |
+
"loss": 0.8701,
|
| 710 |
+
"step": 117000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 70.74,
|
| 714 |
+
"learning_rate": 3.6224000000000004e-05,
|
| 715 |
+
"loss": 0.8692,
|
| 716 |
+
"step": 118000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 71.34,
|
| 720 |
+
"learning_rate": 3.552904347826087e-05,
|
| 721 |
+
"loss": 0.8688,
|
| 722 |
+
"step": 119000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 71.94,
|
| 726 |
+
"learning_rate": 3.483408695652174e-05,
|
| 727 |
+
"loss": 0.8693,
|
| 728 |
+
"step": 120000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 72.54,
|
| 732 |
+
"learning_rate": 3.41384347826087e-05,
|
| 733 |
+
"loss": 0.8675,
|
| 734 |
+
"step": 121000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 73.14,
|
| 738 |
+
"learning_rate": 3.3443478260869565e-05,
|
| 739 |
+
"loss": 0.8667,
|
| 740 |
+
"step": 122000
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 73.74,
|
| 744 |
+
"learning_rate": 3.274852173913044e-05,
|
| 745 |
+
"loss": 0.8667,
|
| 746 |
+
"step": 123000
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 74.34,
|
| 750 |
+
"learning_rate": 3.2052869565217396e-05,
|
| 751 |
+
"loss": 0.8662,
|
| 752 |
+
"step": 124000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 74.94,
|
| 756 |
+
"learning_rate": 3.1357913043478265e-05,
|
| 757 |
+
"loss": 0.8651,
|
| 758 |
+
"step": 125000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 74.94,
|
| 762 |
+
"step": 125000,
|
| 763 |
+
"total_flos": 6.735180748506844e+19,
|
| 764 |
+
"train_loss": 0.06944640869140625,
|
| 765 |
+
"train_runtime": 3177.0339,
|
| 766 |
+
"train_samples_per_second": 80578.303,
|
| 767 |
+
"train_steps_per_second": 39.345
|
| 768 |
+
}
|
| 769 |
+
],
|
| 770 |
+
"max_steps": 125000,
|
| 771 |
+
"num_train_epochs": 75,
|
| 772 |
+
"total_flos": 6.735180748506844e+19,
|
| 773 |
+
"trial_name": null,
|
| 774 |
+
"trial_params": null
|
| 775 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:74561198462778e6b2af69719532ebce91d660f13b77a8a30e5a3ddd5bd948eb
|
| 3 |
+
size 4271
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|