Upload 13 files
Browse files- README.md +45 -1
- all_results.json +15 -0
- config.json +25 -0
- config_backup.json +25 -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 +1147 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
README.md
CHANGED
|
@@ -1,3 +1,47 @@
|
|
| 1 |
---
|
| 2 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
metrics:
|
| 5 |
+
- accuracy
|
| 6 |
+
model-index:
|
| 7 |
+
- name: MLM-final-scraped-g
|
| 8 |
+
results: []
|
| 9 |
---
|
| 10 |
+
|
| 11 |
+
# potocatalysisbert
|
| 12 |
+
|
| 13 |
+
This model is pretrained on a corpus of papers on photcatalysis. For more detailed training procedures, see
|
| 14 |
+
"How beneficial is pre-training on a narrow domain-specific corpus for information extraction about photocatalytic water splitting?" by Taketomo Isazawa and Jacqueline M. Cole.
|
| 15 |
+
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 1.0215
|
| 18 |
+
- Accuracy: 0.7554
|
| 19 |
+
|
| 20 |
+
## Training procedure
|
| 21 |
+
|
| 22 |
+
### Training hyperparameters
|
| 23 |
+
|
| 24 |
+
The following hyperparameters were used during training:
|
| 25 |
+
- learning_rate: 8e-05
|
| 26 |
+
- train_batch_size: 32
|
| 27 |
+
- eval_batch_size: 4
|
| 28 |
+
- seed: 0
|
| 29 |
+
- distributed_type: multi-GPU
|
| 30 |
+
- num_devices: 64
|
| 31 |
+
- total_train_batch_size: 2048
|
| 32 |
+
- total_eval_batch_size: 256
|
| 33 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 34 |
+
- lr_scheduler_type: linear
|
| 35 |
+
- lr_scheduler_warmup_steps: 10000
|
| 36 |
+
- training_steps: 187500
|
| 37 |
+
|
| 38 |
+
### Framework versions
|
| 39 |
+
|
| 40 |
+
- Transformers 4.25.1
|
| 41 |
+
- Pytorch 1.12.0a0+git664058f
|
| 42 |
+
- Datasets 2.7.1
|
| 43 |
+
- Tokenizers 0.12.1
|
| 44 |
+
|
| 45 |
+
## Acknowledgements
|
| 46 |
+
|
| 47 |
+
This model was trained for the paper "How beneficial is pre-training on a narrow domain-specific corpus for information extraction about photocatalytic water splitting?" by Taketomo Isazawa and Jacqueline M. Cole. J.M.C. is grateful for the BASF/Royal Academy of Engineering Research Chair in Data-Driven Molecular Engineering of Functional Materials, which includes PhD studentship support (for T.I.). This Chair is also partly supported by the Science and Technology Facilities Council. They are also indebted to the Argonne Leadership Computing Facility, which is a DOE Office of Science Facility, for use of its research resources, under contract No. DE-AC02-06CH11357.
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 84.96,
|
| 3 |
+
"eval_accuracy": 0.7553893124870495,
|
| 4 |
+
"eval_loss": 1.021484375,
|
| 5 |
+
"eval_runtime": 669.8614,
|
| 6 |
+
"eval_samples": 261426,
|
| 7 |
+
"eval_samples_per_second": 390.269,
|
| 8 |
+
"eval_steps_per_second": 1.526,
|
| 9 |
+
"perplexity": 2.777314281640181,
|
| 10 |
+
"train_loss": 0.10515284895833334,
|
| 11 |
+
"train_runtime": 4947.3494,
|
| 12 |
+
"train_samples": 4518599,
|
| 13 |
+
"train_samples_per_second": 77617.32,
|
| 14 |
+
"train_steps_per_second": 37.899
|
| 15 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/grand/projects/SolarWindowsADSP/taketomo/polaris_outputs/MLM-final-scraped-g/checkpoint-170000",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"torch_dtype": "float16",
|
| 21 |
+
"transformers_version": "4.25.1",
|
| 22 |
+
"type_vocab_size": 2,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 30522
|
| 25 |
+
}
|
config_backup.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/grand/projects/SolarWindowsADSP/taketomo/polaris_outputs/MLM-final-scraped-g/checkpoint-170000",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"torch_dtype": "float16",
|
| 21 |
+
"transformers_version": "4.25.1",
|
| 22 |
+
"type_vocab_size": 2,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 30522
|
| 25 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 84.96,
|
| 3 |
+
"eval_accuracy": 0.7553893124870495,
|
| 4 |
+
"eval_loss": 1.021484375,
|
| 5 |
+
"eval_runtime": 669.8614,
|
| 6 |
+
"eval_samples": 261426,
|
| 7 |
+
"eval_samples_per_second": 390.269,
|
| 8 |
+
"eval_steps_per_second": 1.526,
|
| 9 |
+
"perplexity": 2.777314281640181
|
| 10 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7df55897a38cf818d86199a39d009676b7e4ebdddd92de7fbc9470104a4d0dd9
|
| 3 |
+
size 438085219
|
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": true,
|
| 4 |
+
"mask_token": "[MASK]",
|
| 5 |
+
"model_max_length": 512,
|
| 6 |
+
"name_or_path": "/grand/projects/SolarWindowsADSP/taketomo/polaris_outputs/tokenizer-bert-base-uncased-scraped_only",
|
| 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": 84.96,
|
| 3 |
+
"train_loss": 0.10515284895833334,
|
| 4 |
+
"train_runtime": 4947.3494,
|
| 5 |
+
"train_samples": 4518599,
|
| 6 |
+
"train_samples_per_second": 77617.32,
|
| 7 |
+
"train_steps_per_second": 37.899
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1147 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 84.95695514272768,
|
| 5 |
+
"global_step": 187500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.45,
|
| 12 |
+
"learning_rate": 7.864000000000001e-06,
|
| 13 |
+
"loss": 8.2983,
|
| 14 |
+
"step": 1000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.91,
|
| 18 |
+
"learning_rate": 1.5856e-05,
|
| 19 |
+
"loss": 6.5434,
|
| 20 |
+
"step": 2000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.36,
|
| 24 |
+
"learning_rate": 2.3848e-05,
|
| 25 |
+
"loss": 6.1765,
|
| 26 |
+
"step": 3000
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.81,
|
| 30 |
+
"learning_rate": 3.184000000000001e-05,
|
| 31 |
+
"loss": 5.9601,
|
| 32 |
+
"step": 4000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 2.27,
|
| 36 |
+
"learning_rate": 3.9832e-05,
|
| 37 |
+
"loss": 5.8266,
|
| 38 |
+
"step": 5000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.72,
|
| 42 |
+
"learning_rate": 4.7824e-05,
|
| 43 |
+
"loss": 5.7332,
|
| 44 |
+
"step": 6000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 3.17,
|
| 48 |
+
"learning_rate": 5.5816e-05,
|
| 49 |
+
"loss": 5.6554,
|
| 50 |
+
"step": 7000
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.62,
|
| 54 |
+
"learning_rate": 6.380800000000001e-05,
|
| 55 |
+
"loss": 5.5975,
|
| 56 |
+
"step": 8000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 4.08,
|
| 60 |
+
"learning_rate": 7.18e-05,
|
| 61 |
+
"loss": 5.5568,
|
| 62 |
+
"step": 9000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 4.53,
|
| 66 |
+
"learning_rate": 7.9792e-05,
|
| 67 |
+
"loss": 5.5224,
|
| 68 |
+
"step": 10000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 4.98,
|
| 72 |
+
"learning_rate": 7.95614647887324e-05,
|
| 73 |
+
"loss": 5.4944,
|
| 74 |
+
"step": 11000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 5.44,
|
| 78 |
+
"learning_rate": 7.911121126760565e-05,
|
| 79 |
+
"loss": 5.473,
|
| 80 |
+
"step": 12000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 5.89,
|
| 84 |
+
"learning_rate": 7.866095774647888e-05,
|
| 85 |
+
"loss": 5.4557,
|
| 86 |
+
"step": 13000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 6.34,
|
| 90 |
+
"learning_rate": 7.821070422535212e-05,
|
| 91 |
+
"loss": 5.4392,
|
| 92 |
+
"step": 14000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 6.8,
|
| 96 |
+
"learning_rate": 7.776045070422535e-05,
|
| 97 |
+
"loss": 5.4274,
|
| 98 |
+
"step": 15000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 7.25,
|
| 102 |
+
"learning_rate": 7.73101971830986e-05,
|
| 103 |
+
"loss": 5.4168,
|
| 104 |
+
"step": 16000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 7.7,
|
| 108 |
+
"learning_rate": 7.685994366197184e-05,
|
| 109 |
+
"loss": 5.4051,
|
| 110 |
+
"step": 17000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 8.16,
|
| 114 |
+
"learning_rate": 7.640969014084507e-05,
|
| 115 |
+
"loss": 5.3963,
|
| 116 |
+
"step": 18000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 8.61,
|
| 120 |
+
"learning_rate": 7.595943661971832e-05,
|
| 121 |
+
"loss": 5.3901,
|
| 122 |
+
"step": 19000
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 9.06,
|
| 126 |
+
"learning_rate": 7.550918309859156e-05,
|
| 127 |
+
"loss": 5.3848,
|
| 128 |
+
"step": 20000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 9.52,
|
| 132 |
+
"learning_rate": 7.505892957746479e-05,
|
| 133 |
+
"loss": 5.2904,
|
| 134 |
+
"step": 21000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 9.97,
|
| 138 |
+
"learning_rate": 7.460867605633804e-05,
|
| 139 |
+
"loss": 4.6695,
|
| 140 |
+
"step": 22000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 10.42,
|
| 144 |
+
"learning_rate": 7.415842253521126e-05,
|
| 145 |
+
"loss": 3.6226,
|
| 146 |
+
"step": 23000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 10.87,
|
| 150 |
+
"learning_rate": 7.370816901408451e-05,
|
| 151 |
+
"loss": 2.5281,
|
| 152 |
+
"step": 24000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 11.33,
|
| 156 |
+
"learning_rate": 7.325791549295775e-05,
|
| 157 |
+
"loss": 2.2386,
|
| 158 |
+
"step": 25000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 11.78,
|
| 162 |
+
"learning_rate": 7.2807661971831e-05,
|
| 163 |
+
"loss": 2.0823,
|
| 164 |
+
"step": 26000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 12.23,
|
| 168 |
+
"learning_rate": 7.235740845070423e-05,
|
| 169 |
+
"loss": 1.9728,
|
| 170 |
+
"step": 27000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 12.69,
|
| 174 |
+
"learning_rate": 7.190715492957747e-05,
|
| 175 |
+
"loss": 1.8923,
|
| 176 |
+
"step": 28000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 13.14,
|
| 180 |
+
"learning_rate": 7.145690140845072e-05,
|
| 181 |
+
"loss": 1.8241,
|
| 182 |
+
"step": 29000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 13.59,
|
| 186 |
+
"learning_rate": 7.100664788732395e-05,
|
| 187 |
+
"loss": 1.7682,
|
| 188 |
+
"step": 30000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 14.05,
|
| 192 |
+
"learning_rate": 7.055639436619719e-05,
|
| 193 |
+
"loss": 1.7244,
|
| 194 |
+
"step": 31000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 14.5,
|
| 198 |
+
"learning_rate": 7.010614084507043e-05,
|
| 199 |
+
"loss": 1.6804,
|
| 200 |
+
"step": 32000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 14.95,
|
| 204 |
+
"learning_rate": 6.965588732394366e-05,
|
| 205 |
+
"loss": 1.6465,
|
| 206 |
+
"step": 33000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 15.41,
|
| 210 |
+
"learning_rate": 6.920563380281691e-05,
|
| 211 |
+
"loss": 1.6159,
|
| 212 |
+
"step": 34000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 15.86,
|
| 216 |
+
"learning_rate": 6.875538028169015e-05,
|
| 217 |
+
"loss": 1.5909,
|
| 218 |
+
"step": 35000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 16.31,
|
| 222 |
+
"learning_rate": 6.830512676056338e-05,
|
| 223 |
+
"loss": 1.5664,
|
| 224 |
+
"step": 36000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 16.76,
|
| 228 |
+
"learning_rate": 6.785487323943663e-05,
|
| 229 |
+
"loss": 1.5481,
|
| 230 |
+
"step": 37000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 17.22,
|
| 234 |
+
"learning_rate": 6.740461971830987e-05,
|
| 235 |
+
"loss": 1.5282,
|
| 236 |
+
"step": 38000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 17.67,
|
| 240 |
+
"learning_rate": 6.69543661971831e-05,
|
| 241 |
+
"loss": 1.5103,
|
| 242 |
+
"step": 39000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 18.12,
|
| 246 |
+
"learning_rate": 6.650411267605634e-05,
|
| 247 |
+
"loss": 1.4946,
|
| 248 |
+
"step": 40000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 18.58,
|
| 252 |
+
"learning_rate": 6.605385915492959e-05,
|
| 253 |
+
"loss": 1.4793,
|
| 254 |
+
"step": 41000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 19.03,
|
| 258 |
+
"learning_rate": 6.560360563380282e-05,
|
| 259 |
+
"loss": 1.4647,
|
| 260 |
+
"step": 42000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 19.48,
|
| 264 |
+
"learning_rate": 6.515335211267606e-05,
|
| 265 |
+
"loss": 1.4525,
|
| 266 |
+
"step": 43000
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 19.94,
|
| 270 |
+
"learning_rate": 6.47030985915493e-05,
|
| 271 |
+
"loss": 1.4408,
|
| 272 |
+
"step": 44000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 20.39,
|
| 276 |
+
"learning_rate": 6.425284507042254e-05,
|
| 277 |
+
"loss": 1.429,
|
| 278 |
+
"step": 45000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 20.84,
|
| 282 |
+
"learning_rate": 6.380259154929578e-05,
|
| 283 |
+
"loss": 1.4191,
|
| 284 |
+
"step": 46000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 21.3,
|
| 288 |
+
"learning_rate": 6.335233802816903e-05,
|
| 289 |
+
"loss": 1.4093,
|
| 290 |
+
"step": 47000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 21.75,
|
| 294 |
+
"learning_rate": 6.290208450704226e-05,
|
| 295 |
+
"loss": 1.399,
|
| 296 |
+
"step": 48000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 22.2,
|
| 300 |
+
"learning_rate": 6.24518309859155e-05,
|
| 301 |
+
"loss": 1.3902,
|
| 302 |
+
"step": 49000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 22.66,
|
| 306 |
+
"learning_rate": 6.200157746478873e-05,
|
| 307 |
+
"loss": 1.3811,
|
| 308 |
+
"step": 50000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 23.11,
|
| 312 |
+
"learning_rate": 6.155132394366198e-05,
|
| 313 |
+
"loss": 1.3738,
|
| 314 |
+
"step": 51000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 23.56,
|
| 318 |
+
"learning_rate": 6.110107042253522e-05,
|
| 319 |
+
"loss": 1.3651,
|
| 320 |
+
"step": 52000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 24.01,
|
| 324 |
+
"learning_rate": 6.0650816901408453e-05,
|
| 325 |
+
"loss": 1.361,
|
| 326 |
+
"step": 53000
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 24.47,
|
| 330 |
+
"learning_rate": 6.0200563380281696e-05,
|
| 331 |
+
"loss": 1.3519,
|
| 332 |
+
"step": 54000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 24.92,
|
| 336 |
+
"learning_rate": 5.975030985915493e-05,
|
| 337 |
+
"loss": 1.3456,
|
| 338 |
+
"step": 55000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 25.37,
|
| 342 |
+
"learning_rate": 5.9300056338028174e-05,
|
| 343 |
+
"loss": 1.3394,
|
| 344 |
+
"step": 56000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 25.83,
|
| 348 |
+
"learning_rate": 5.8849802816901416e-05,
|
| 349 |
+
"loss": 1.3332,
|
| 350 |
+
"step": 57000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 26.28,
|
| 354 |
+
"learning_rate": 5.839954929577465e-05,
|
| 355 |
+
"loss": 1.3273,
|
| 356 |
+
"step": 58000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 26.73,
|
| 360 |
+
"learning_rate": 5.7949295774647894e-05,
|
| 361 |
+
"loss": 1.3222,
|
| 362 |
+
"step": 59000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 27.19,
|
| 366 |
+
"learning_rate": 5.7499042253521136e-05,
|
| 367 |
+
"loss": 1.3159,
|
| 368 |
+
"step": 60000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 27.64,
|
| 372 |
+
"learning_rate": 5.7048788732394365e-05,
|
| 373 |
+
"loss": 1.3117,
|
| 374 |
+
"step": 61000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 28.09,
|
| 378 |
+
"learning_rate": 5.659853521126761e-05,
|
| 379 |
+
"loss": 1.3067,
|
| 380 |
+
"step": 62000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 28.55,
|
| 384 |
+
"learning_rate": 5.614828169014085e-05,
|
| 385 |
+
"loss": 1.3015,
|
| 386 |
+
"step": 63000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 29.0,
|
| 390 |
+
"learning_rate": 5.569802816901409e-05,
|
| 391 |
+
"loss": 1.2972,
|
| 392 |
+
"step": 64000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 29.45,
|
| 396 |
+
"learning_rate": 5.524777464788733e-05,
|
| 397 |
+
"loss": 1.291,
|
| 398 |
+
"step": 65000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 29.9,
|
| 402 |
+
"learning_rate": 5.479752112676057e-05,
|
| 403 |
+
"loss": 1.2877,
|
| 404 |
+
"step": 66000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 30.36,
|
| 408 |
+
"learning_rate": 5.434726760563381e-05,
|
| 409 |
+
"loss": 1.2836,
|
| 410 |
+
"step": 67000
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 30.81,
|
| 414 |
+
"learning_rate": 5.389701408450704e-05,
|
| 415 |
+
"loss": 1.2798,
|
| 416 |
+
"step": 68000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 31.26,
|
| 420 |
+
"learning_rate": 5.344721126760564e-05,
|
| 421 |
+
"loss": 1.276,
|
| 422 |
+
"step": 69000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 31.72,
|
| 426 |
+
"learning_rate": 5.2996507042253526e-05,
|
| 427 |
+
"loss": 1.2723,
|
| 428 |
+
"step": 70000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 32.17,
|
| 432 |
+
"learning_rate": 5.254670422535211e-05,
|
| 433 |
+
"loss": 1.2676,
|
| 434 |
+
"step": 71000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 32.62,
|
| 438 |
+
"learning_rate": 5.2096000000000004e-05,
|
| 439 |
+
"loss": 1.2645,
|
| 440 |
+
"step": 72000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 33.08,
|
| 444 |
+
"learning_rate": 5.1645746478873246e-05,
|
| 445 |
+
"loss": 1.2604,
|
| 446 |
+
"step": 73000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 33.53,
|
| 450 |
+
"learning_rate": 5.119549295774648e-05,
|
| 451 |
+
"loss": 1.2577,
|
| 452 |
+
"step": 74000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 33.98,
|
| 456 |
+
"learning_rate": 5.0745239436619724e-05,
|
| 457 |
+
"loss": 1.2552,
|
| 458 |
+
"step": 75000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 34.44,
|
| 462 |
+
"learning_rate": 5.029498591549297e-05,
|
| 463 |
+
"loss": 1.2516,
|
| 464 |
+
"step": 76000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 34.89,
|
| 468 |
+
"learning_rate": 4.9844732394366195e-05,
|
| 469 |
+
"loss": 1.2488,
|
| 470 |
+
"step": 77000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 35.34,
|
| 474 |
+
"learning_rate": 4.939447887323944e-05,
|
| 475 |
+
"loss": 1.2458,
|
| 476 |
+
"step": 78000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 35.8,
|
| 480 |
+
"learning_rate": 4.894467605633804e-05,
|
| 481 |
+
"loss": 1.2443,
|
| 482 |
+
"step": 79000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 36.25,
|
| 486 |
+
"learning_rate": 4.8493971830985916e-05,
|
| 487 |
+
"loss": 1.2398,
|
| 488 |
+
"step": 80000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 36.7,
|
| 492 |
+
"learning_rate": 4.804371830985916e-05,
|
| 493 |
+
"loss": 1.2374,
|
| 494 |
+
"step": 81000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 37.15,
|
| 498 |
+
"learning_rate": 4.75934647887324e-05,
|
| 499 |
+
"loss": 1.2341,
|
| 500 |
+
"step": 82000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 37.61,
|
| 504 |
+
"learning_rate": 4.714321126760564e-05,
|
| 505 |
+
"loss": 1.2321,
|
| 506 |
+
"step": 83000
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 38.06,
|
| 510 |
+
"learning_rate": 4.669295774647888e-05,
|
| 511 |
+
"loss": 1.23,
|
| 512 |
+
"step": 84000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 38.51,
|
| 516 |
+
"learning_rate": 4.624270422535212e-05,
|
| 517 |
+
"loss": 1.2265,
|
| 518 |
+
"step": 85000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 38.97,
|
| 522 |
+
"learning_rate": 4.5792450704225356e-05,
|
| 523 |
+
"loss": 1.2253,
|
| 524 |
+
"step": 86000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 39.42,
|
| 528 |
+
"learning_rate": 4.534219718309859e-05,
|
| 529 |
+
"loss": 1.2227,
|
| 530 |
+
"step": 87000
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 39.87,
|
| 534 |
+
"learning_rate": 4.4891943661971834e-05,
|
| 535 |
+
"loss": 1.2204,
|
| 536 |
+
"step": 88000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 40.33,
|
| 540 |
+
"learning_rate": 4.4441690140845077e-05,
|
| 541 |
+
"loss": 1.2171,
|
| 542 |
+
"step": 89000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 40.78,
|
| 546 |
+
"learning_rate": 4.399143661971831e-05,
|
| 547 |
+
"loss": 1.2154,
|
| 548 |
+
"step": 90000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 41.23,
|
| 552 |
+
"learning_rate": 4.3541183098591555e-05,
|
| 553 |
+
"loss": 1.2132,
|
| 554 |
+
"step": 91000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 41.69,
|
| 558 |
+
"learning_rate": 4.30909295774648e-05,
|
| 559 |
+
"loss": 1.2118,
|
| 560 |
+
"step": 92000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 42.14,
|
| 564 |
+
"learning_rate": 4.2640676056338026e-05,
|
| 565 |
+
"loss": 1.2088,
|
| 566 |
+
"step": 93000
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 42.59,
|
| 570 |
+
"learning_rate": 4.2189971830985916e-05,
|
| 571 |
+
"loss": 1.2068,
|
| 572 |
+
"step": 94000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 43.04,
|
| 576 |
+
"learning_rate": 4.173971830985916e-05,
|
| 577 |
+
"loss": 1.2063,
|
| 578 |
+
"step": 95000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 43.5,
|
| 582 |
+
"learning_rate": 4.12894647887324e-05,
|
| 583 |
+
"loss": 1.2025,
|
| 584 |
+
"step": 96000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 43.95,
|
| 588 |
+
"learning_rate": 4.083921126760564e-05,
|
| 589 |
+
"loss": 1.2013,
|
| 590 |
+
"step": 97000
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 44.4,
|
| 594 |
+
"learning_rate": 4.038895774647888e-05,
|
| 595 |
+
"loss": 1.2002,
|
| 596 |
+
"step": 98000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 44.86,
|
| 600 |
+
"learning_rate": 3.993825352112676e-05,
|
| 601 |
+
"loss": 1.198,
|
| 602 |
+
"step": 99000
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 45.31,
|
| 606 |
+
"learning_rate": 3.9488e-05,
|
| 607 |
+
"loss": 1.1962,
|
| 608 |
+
"step": 100000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 45.76,
|
| 612 |
+
"learning_rate": 3.903774647887324e-05,
|
| 613 |
+
"loss": 1.1953,
|
| 614 |
+
"step": 101000
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 46.22,
|
| 618 |
+
"learning_rate": 3.8587492957746483e-05,
|
| 619 |
+
"loss": 1.1927,
|
| 620 |
+
"step": 102000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 46.67,
|
| 624 |
+
"learning_rate": 3.813723943661972e-05,
|
| 625 |
+
"loss": 1.1909,
|
| 626 |
+
"step": 103000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 47.12,
|
| 630 |
+
"learning_rate": 3.768698591549296e-05,
|
| 631 |
+
"loss": 1.1891,
|
| 632 |
+
"step": 104000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 47.58,
|
| 636 |
+
"learning_rate": 3.72367323943662e-05,
|
| 637 |
+
"loss": 1.1876,
|
| 638 |
+
"step": 105000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 48.03,
|
| 642 |
+
"learning_rate": 3.678647887323944e-05,
|
| 643 |
+
"loss": 1.1868,
|
| 644 |
+
"step": 106000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 48.48,
|
| 648 |
+
"learning_rate": 3.633622535211268e-05,
|
| 649 |
+
"loss": 1.1841,
|
| 650 |
+
"step": 107000
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 48.94,
|
| 654 |
+
"learning_rate": 3.588597183098592e-05,
|
| 655 |
+
"loss": 1.1844,
|
| 656 |
+
"step": 108000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 49.39,
|
| 660 |
+
"learning_rate": 3.543571830985916e-05,
|
| 661 |
+
"loss": 1.1815,
|
| 662 |
+
"step": 109000
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 49.84,
|
| 666 |
+
"learning_rate": 3.4985464788732395e-05,
|
| 667 |
+
"loss": 1.1809,
|
| 668 |
+
"step": 110000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 50.29,
|
| 672 |
+
"learning_rate": 3.453521126760564e-05,
|
| 673 |
+
"loss": 1.1784,
|
| 674 |
+
"step": 111000
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 50.75,
|
| 678 |
+
"learning_rate": 3.408495774647888e-05,
|
| 679 |
+
"loss": 1.1774,
|
| 680 |
+
"step": 112000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 51.2,
|
| 684 |
+
"learning_rate": 3.3634704225352115e-05,
|
| 685 |
+
"loss": 1.1761,
|
| 686 |
+
"step": 113000
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 51.65,
|
| 690 |
+
"learning_rate": 3.318445070422535e-05,
|
| 691 |
+
"loss": 1.175,
|
| 692 |
+
"step": 114000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 52.11,
|
| 696 |
+
"learning_rate": 3.273419718309859e-05,
|
| 697 |
+
"loss": 1.1742,
|
| 698 |
+
"step": 115000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 52.56,
|
| 702 |
+
"learning_rate": 3.2283943661971836e-05,
|
| 703 |
+
"loss": 1.1726,
|
| 704 |
+
"step": 116000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 53.01,
|
| 708 |
+
"learning_rate": 3.183323943661972e-05,
|
| 709 |
+
"loss": 1.1719,
|
| 710 |
+
"step": 117000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 53.47,
|
| 714 |
+
"learning_rate": 3.1382985915492955e-05,
|
| 715 |
+
"loss": 1.1699,
|
| 716 |
+
"step": 118000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 53.92,
|
| 720 |
+
"learning_rate": 3.09327323943662e-05,
|
| 721 |
+
"loss": 1.1684,
|
| 722 |
+
"step": 119000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 54.37,
|
| 726 |
+
"learning_rate": 3.048247887323944e-05,
|
| 727 |
+
"loss": 1.1676,
|
| 728 |
+
"step": 120000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 54.83,
|
| 732 |
+
"learning_rate": 3.003222535211268e-05,
|
| 733 |
+
"loss": 1.1662,
|
| 734 |
+
"step": 121000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 55.28,
|
| 738 |
+
"learning_rate": 2.9581971830985918e-05,
|
| 739 |
+
"loss": 1.1647,
|
| 740 |
+
"step": 122000
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 55.73,
|
| 744 |
+
"learning_rate": 2.9131718309859157e-05,
|
| 745 |
+
"loss": 1.1643,
|
| 746 |
+
"step": 123000
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 56.18,
|
| 750 |
+
"learning_rate": 2.86814647887324e-05,
|
| 751 |
+
"loss": 1.1639,
|
| 752 |
+
"step": 124000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 56.64,
|
| 756 |
+
"learning_rate": 2.8231211267605635e-05,
|
| 757 |
+
"loss": 1.1617,
|
| 758 |
+
"step": 125000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 57.09,
|
| 762 |
+
"learning_rate": 2.7780957746478874e-05,
|
| 763 |
+
"loss": 1.1606,
|
| 764 |
+
"step": 126000
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 57.54,
|
| 768 |
+
"learning_rate": 2.7330704225352116e-05,
|
| 769 |
+
"loss": 1.1608,
|
| 770 |
+
"step": 127000
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 58.0,
|
| 774 |
+
"learning_rate": 2.6880450704225355e-05,
|
| 775 |
+
"loss": 1.159,
|
| 776 |
+
"step": 128000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 58.45,
|
| 780 |
+
"learning_rate": 2.643019718309859e-05,
|
| 781 |
+
"loss": 1.1572,
|
| 782 |
+
"step": 129000
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 58.9,
|
| 786 |
+
"learning_rate": 2.5979943661971833e-05,
|
| 787 |
+
"loss": 1.1572,
|
| 788 |
+
"step": 130000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 59.36,
|
| 792 |
+
"learning_rate": 2.5529690140845072e-05,
|
| 793 |
+
"loss": 1.1562,
|
| 794 |
+
"step": 131000
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 59.81,
|
| 798 |
+
"learning_rate": 2.5079436619718314e-05,
|
| 799 |
+
"loss": 1.1551,
|
| 800 |
+
"step": 132000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 60.26,
|
| 804 |
+
"learning_rate": 2.462918309859155e-05,
|
| 805 |
+
"loss": 1.1547,
|
| 806 |
+
"step": 133000
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 60.72,
|
| 810 |
+
"learning_rate": 2.417892957746479e-05,
|
| 811 |
+
"loss": 1.153,
|
| 812 |
+
"step": 134000
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 61.17,
|
| 816 |
+
"learning_rate": 2.372867605633803e-05,
|
| 817 |
+
"loss": 1.1526,
|
| 818 |
+
"step": 135000
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 61.62,
|
| 822 |
+
"learning_rate": 2.327842253521127e-05,
|
| 823 |
+
"loss": 1.1522,
|
| 824 |
+
"step": 136000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 62.08,
|
| 828 |
+
"learning_rate": 2.2828169014084506e-05,
|
| 829 |
+
"loss": 1.1505,
|
| 830 |
+
"step": 137000
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 62.53,
|
| 834 |
+
"learning_rate": 2.2377915492957748e-05,
|
| 835 |
+
"loss": 1.1505,
|
| 836 |
+
"step": 138000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 62.98,
|
| 840 |
+
"learning_rate": 2.1927661971830987e-05,
|
| 841 |
+
"loss": 1.1485,
|
| 842 |
+
"step": 139000
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 63.43,
|
| 846 |
+
"learning_rate": 2.1477859154929578e-05,
|
| 847 |
+
"loss": 1.1482,
|
| 848 |
+
"step": 140000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 63.89,
|
| 852 |
+
"learning_rate": 2.102715492957747e-05,
|
| 853 |
+
"loss": 1.1466,
|
| 854 |
+
"step": 141000
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 64.34,
|
| 858 |
+
"learning_rate": 2.0576901408450704e-05,
|
| 859 |
+
"loss": 1.1467,
|
| 860 |
+
"step": 142000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 64.79,
|
| 864 |
+
"learning_rate": 2.0126647887323946e-05,
|
| 865 |
+
"loss": 1.146,
|
| 866 |
+
"step": 143000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 65.25,
|
| 870 |
+
"learning_rate": 1.9676394366197185e-05,
|
| 871 |
+
"loss": 1.1452,
|
| 872 |
+
"step": 144000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 65.7,
|
| 876 |
+
"learning_rate": 1.9226140845070424e-05,
|
| 877 |
+
"loss": 1.1447,
|
| 878 |
+
"step": 145000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 66.15,
|
| 882 |
+
"learning_rate": 1.8775887323943663e-05,
|
| 883 |
+
"loss": 1.1441,
|
| 884 |
+
"step": 146000
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 66.61,
|
| 888 |
+
"learning_rate": 1.8325633802816902e-05,
|
| 889 |
+
"loss": 1.143,
|
| 890 |
+
"step": 147000
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 67.06,
|
| 894 |
+
"learning_rate": 1.787538028169014e-05,
|
| 895 |
+
"loss": 1.1433,
|
| 896 |
+
"step": 148000
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 67.51,
|
| 900 |
+
"learning_rate": 1.7425126760563384e-05,
|
| 901 |
+
"loss": 1.1414,
|
| 902 |
+
"step": 149000
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 67.97,
|
| 906 |
+
"learning_rate": 1.697487323943662e-05,
|
| 907 |
+
"loss": 1.1413,
|
| 908 |
+
"step": 150000
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 68.42,
|
| 912 |
+
"learning_rate": 1.652461971830986e-05,
|
| 913 |
+
"loss": 1.14,
|
| 914 |
+
"step": 151000
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 68.87,
|
| 918 |
+
"learning_rate": 1.60743661971831e-05,
|
| 919 |
+
"loss": 1.14,
|
| 920 |
+
"step": 152000
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 69.32,
|
| 924 |
+
"learning_rate": 1.562411267605634e-05,
|
| 925 |
+
"loss": 1.139,
|
| 926 |
+
"step": 153000
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 69.78,
|
| 930 |
+
"learning_rate": 1.5173859154929578e-05,
|
| 931 |
+
"loss": 1.1385,
|
| 932 |
+
"step": 154000
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 70.23,
|
| 936 |
+
"learning_rate": 1.4723605633802817e-05,
|
| 937 |
+
"loss": 1.1385,
|
| 938 |
+
"step": 155000
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 70.68,
|
| 942 |
+
"learning_rate": 1.4272901408450706e-05,
|
| 943 |
+
"loss": 1.1374,
|
| 944 |
+
"step": 156000
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 71.14,
|
| 948 |
+
"learning_rate": 1.3822647887323945e-05,
|
| 949 |
+
"loss": 1.1362,
|
| 950 |
+
"step": 157000
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 71.59,
|
| 954 |
+
"learning_rate": 1.3372394366197183e-05,
|
| 955 |
+
"loss": 1.137,
|
| 956 |
+
"step": 158000
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 72.04,
|
| 960 |
+
"learning_rate": 1.2922140845070423e-05,
|
| 961 |
+
"loss": 1.1355,
|
| 962 |
+
"step": 159000
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 72.5,
|
| 966 |
+
"learning_rate": 1.2471887323943664e-05,
|
| 967 |
+
"loss": 1.1342,
|
| 968 |
+
"step": 160000
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 72.95,
|
| 972 |
+
"learning_rate": 1.2021633802816903e-05,
|
| 973 |
+
"loss": 1.1347,
|
| 974 |
+
"step": 161000
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 73.4,
|
| 978 |
+
"learning_rate": 1.157138028169014e-05,
|
| 979 |
+
"loss": 1.133,
|
| 980 |
+
"step": 162000
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 73.86,
|
| 984 |
+
"learning_rate": 1.1121126760563381e-05,
|
| 985 |
+
"loss": 1.1339,
|
| 986 |
+
"step": 163000
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 74.31,
|
| 990 |
+
"learning_rate": 1.0670873239436622e-05,
|
| 991 |
+
"loss": 1.1332,
|
| 992 |
+
"step": 164000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 74.76,
|
| 996 |
+
"learning_rate": 1.022061971830986e-05,
|
| 997 |
+
"loss": 1.1319,
|
| 998 |
+
"step": 165000
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 75.22,
|
| 1002 |
+
"learning_rate": 9.7703661971831e-06,
|
| 1003 |
+
"loss": 1.1311,
|
| 1004 |
+
"step": 166000
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 75.67,
|
| 1008 |
+
"learning_rate": 9.32056338028169e-06,
|
| 1009 |
+
"loss": 1.1303,
|
| 1010 |
+
"step": 167000
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 76.12,
|
| 1014 |
+
"learning_rate": 8.869859154929579e-06,
|
| 1015 |
+
"loss": 1.1318,
|
| 1016 |
+
"step": 168000
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 76.57,
|
| 1020 |
+
"learning_rate": 8.419605633802818e-06,
|
| 1021 |
+
"loss": 1.1304,
|
| 1022 |
+
"step": 169000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 77.03,
|
| 1026 |
+
"learning_rate": 7.969352112676057e-06,
|
| 1027 |
+
"loss": 1.1297,
|
| 1028 |
+
"step": 170000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 77.48,
|
| 1032 |
+
"learning_rate": 7.519098591549297e-06,
|
| 1033 |
+
"loss": 1.1295,
|
| 1034 |
+
"step": 171000
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 77.93,
|
| 1038 |
+
"learning_rate": 7.068845070422535e-06,
|
| 1039 |
+
"loss": 1.1291,
|
| 1040 |
+
"step": 172000
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 78.39,
|
| 1044 |
+
"learning_rate": 6.618591549295776e-06,
|
| 1045 |
+
"loss": 1.1287,
|
| 1046 |
+
"step": 173000
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 78.84,
|
| 1050 |
+
"learning_rate": 6.167887323943662e-06,
|
| 1051 |
+
"loss": 1.128,
|
| 1052 |
+
"step": 174000
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 79.29,
|
| 1056 |
+
"learning_rate": 5.717633802816902e-06,
|
| 1057 |
+
"loss": 1.1283,
|
| 1058 |
+
"step": 175000
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 79.75,
|
| 1062 |
+
"learning_rate": 5.267380281690141e-06,
|
| 1063 |
+
"loss": 1.1272,
|
| 1064 |
+
"step": 176000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 80.2,
|
| 1068 |
+
"learning_rate": 4.817126760563381e-06,
|
| 1069 |
+
"loss": 1.1273,
|
| 1070 |
+
"step": 177000
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 80.65,
|
| 1074 |
+
"learning_rate": 4.366873239436621e-06,
|
| 1075 |
+
"loss": 1.1271,
|
| 1076 |
+
"step": 178000
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 81.11,
|
| 1080 |
+
"learning_rate": 3.9166197183098595e-06,
|
| 1081 |
+
"loss": 1.126,
|
| 1082 |
+
"step": 179000
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 81.56,
|
| 1086 |
+
"learning_rate": 3.4663661971830985e-06,
|
| 1087 |
+
"loss": 1.1263,
|
| 1088 |
+
"step": 180000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 82.01,
|
| 1092 |
+
"learning_rate": 3.016112676056338e-06,
|
| 1093 |
+
"loss": 1.1261,
|
| 1094 |
+
"step": 181000
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 82.46,
|
| 1098 |
+
"learning_rate": 2.5658591549295773e-06,
|
| 1099 |
+
"loss": 1.1247,
|
| 1100 |
+
"step": 182000
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 82.92,
|
| 1104 |
+
"learning_rate": 2.115605633802817e-06,
|
| 1105 |
+
"loss": 1.1267,
|
| 1106 |
+
"step": 183000
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 83.37,
|
| 1110 |
+
"learning_rate": 1.6653521126760563e-06,
|
| 1111 |
+
"loss": 1.1246,
|
| 1112 |
+
"step": 184000
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 83.82,
|
| 1116 |
+
"learning_rate": 1.2150985915492959e-06,
|
| 1117 |
+
"loss": 1.1251,
|
| 1118 |
+
"step": 185000
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 84.28,
|
| 1122 |
+
"learning_rate": 7.648450704225354e-07,
|
| 1123 |
+
"loss": 1.1251,
|
| 1124 |
+
"step": 186000
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 84.73,
|
| 1128 |
+
"learning_rate": 3.145915492957747e-07,
|
| 1129 |
+
"loss": 1.1246,
|
| 1130 |
+
"step": 187000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 84.96,
|
| 1134 |
+
"step": 187500,
|
| 1135 |
+
"total_flos": 1.0104234636302352e+20,
|
| 1136 |
+
"train_loss": 0.10515284895833334,
|
| 1137 |
+
"train_runtime": 4947.3494,
|
| 1138 |
+
"train_samples_per_second": 77617.32,
|
| 1139 |
+
"train_steps_per_second": 37.899
|
| 1140 |
+
}
|
| 1141 |
+
],
|
| 1142 |
+
"max_steps": 187500,
|
| 1143 |
+
"num_train_epochs": 85,
|
| 1144 |
+
"total_flos": 1.0104234636302352e+20,
|
| 1145 |
+
"trial_name": null,
|
| 1146 |
+
"trial_params": null
|
| 1147 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebf1380be2a758a012ae4f84303799c4f081a889cdd1c60898e1584a869d0fc6
|
| 3 |
+
size 4271
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|