gokuls commited on
Commit
8e63567
1 Parent(s): 281daee

End of training

Browse files
Files changed (5) hide show
  1. README.md +19 -2
  2. all_results.json +15 -0
  3. eval_results.json +10 -0
  4. train_results.json +8 -0
  5. trainer_state.json +3538 -0
README.md CHANGED
@@ -1,9 +1,23 @@
1
  ---
2
  tags:
3
  - generated_from_trainer
 
 
 
 
4
  model-index:
5
  - name: HBERTv1_L12_H768_A12_ffn_1
6
- results: []
 
 
 
 
 
 
 
 
 
 
7
  ---
8
 
9
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -11,7 +25,10 @@ should probably proofread and complete it, then remove this comment. -->
11
 
12
  # HBERTv1_L12_H768_A12_ffn_1
13
 
14
- This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
 
 
 
15
 
16
  ## Model description
17
 
 
1
  ---
2
  tags:
3
  - generated_from_trainer
4
+ datasets:
5
+ - gokuls/wiki_book_corpus_complete_processed_bert_dataset
6
+ metrics:
7
+ - accuracy
8
  model-index:
9
  - name: HBERTv1_L12_H768_A12_ffn_1
10
+ results:
11
+ - task:
12
+ name: Masked Language Modeling
13
+ type: fill-mask
14
+ dataset:
15
+ name: gokuls/wiki_book_corpus_complete_processed_bert_dataset
16
+ type: gokuls/wiki_book_corpus_complete_processed_bert_dataset
17
+ metrics:
18
+ - name: Accuracy
19
+ type: accuracy
20
+ value: 0.4973171408449596
21
  ---
22
 
23
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
25
 
26
  # HBERTv1_L12_H768_A12_ffn_1
27
 
28
+ This model is a fine-tuned version of [](https://huggingface.co/) on the gokuls/wiki_book_corpus_complete_processed_bert_dataset dataset.
29
+ It achieves the following results on the evaluation set:
30
+ - Loss: 2.8656
31
+ - Accuracy: 0.4973
32
 
33
  ## Model description
34
 
all_results.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.4,
3
+ "eval_accuracy": 0.4973171408449596,
4
+ "eval_loss": 2.8656117916107178,
5
+ "eval_runtime": 1575.6817,
6
+ "eval_samples": 308356,
7
+ "eval_samples_per_second": 195.697,
8
+ "eval_steps_per_second": 4.078,
9
+ "perplexity": 17.559792852991617,
10
+ "train_loss": 4.907397945397096,
11
+ "train_runtime": 197998.9762,
12
+ "train_samples": 5858758,
13
+ "train_samples_per_second": 2958.984,
14
+ "train_steps_per_second": 61.646
15
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.4,
3
+ "eval_accuracy": 0.4973171408449596,
4
+ "eval_loss": 2.8656117916107178,
5
+ "eval_runtime": 1575.6817,
6
+ "eval_samples": 308356,
7
+ "eval_samples_per_second": 195.697,
8
+ "eval_steps_per_second": 4.078,
9
+ "perplexity": 17.559792852991617
10
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.4,
3
+ "train_loss": 4.907397945397096,
4
+ "train_runtime": 197998.9762,
5
+ "train_samples": 5858758,
6
+ "train_samples_per_second": 2958.984,
7
+ "train_steps_per_second": 61.646
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,3538 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.399203657277688,
5
+ "eval_steps": 1000000,
6
+ "global_step": 292842,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 5.000000000000001e-07,
14
+ "loss": 9.9527,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 1.0000000000000002e-06,
20
+ "loss": 9.1966,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 1.5e-06,
26
+ "loss": 8.7737,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.02,
31
+ "learning_rate": 2.0000000000000003e-06,
32
+ "loss": 8.3012,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.02,
37
+ "learning_rate": 2.5e-06,
38
+ "loss": 7.8046,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.02,
43
+ "learning_rate": 3e-06,
44
+ "loss": 7.4149,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.03,
49
+ "learning_rate": 3.5e-06,
50
+ "loss": 7.1666,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.03,
55
+ "learning_rate": 4.000000000000001e-06,
56
+ "loss": 7.0257,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.04,
61
+ "learning_rate": 4.5e-06,
62
+ "loss": 6.9423,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.04,
67
+ "learning_rate": 5e-06,
68
+ "loss": 6.8754,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.05,
73
+ "learning_rate": 5.500000000000001e-06,
74
+ "loss": 6.8221,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.05,
79
+ "learning_rate": 6e-06,
80
+ "loss": 6.775,
81
+ "step": 6000
82
+ },
83
+ {
84
+ "epoch": 0.05,
85
+ "learning_rate": 6.5000000000000004e-06,
86
+ "loss": 6.7376,
87
+ "step": 6500
88
+ },
89
+ {
90
+ "epoch": 0.06,
91
+ "learning_rate": 7e-06,
92
+ "loss": 6.7071,
93
+ "step": 7000
94
+ },
95
+ {
96
+ "epoch": 0.06,
97
+ "learning_rate": 7.500000000000001e-06,
98
+ "loss": 6.6743,
99
+ "step": 7500
100
+ },
101
+ {
102
+ "epoch": 0.07,
103
+ "learning_rate": 8.000000000000001e-06,
104
+ "loss": 6.639,
105
+ "step": 8000
106
+ },
107
+ {
108
+ "epoch": 0.07,
109
+ "learning_rate": 8.5e-06,
110
+ "loss": 6.6047,
111
+ "step": 8500
112
+ },
113
+ {
114
+ "epoch": 0.07,
115
+ "learning_rate": 9e-06,
116
+ "loss": 6.5799,
117
+ "step": 9000
118
+ },
119
+ {
120
+ "epoch": 0.08,
121
+ "learning_rate": 9.5e-06,
122
+ "loss": 6.5529,
123
+ "step": 9500
124
+ },
125
+ {
126
+ "epoch": 0.08,
127
+ "learning_rate": 1e-05,
128
+ "loss": 6.5341,
129
+ "step": 10000
130
+ },
131
+ {
132
+ "epoch": 0.09,
133
+ "learning_rate": 9.999590022794733e-06,
134
+ "loss": 6.5078,
135
+ "step": 10500
136
+ },
137
+ {
138
+ "epoch": 0.09,
139
+ "learning_rate": 9.999180045589466e-06,
140
+ "loss": 6.4779,
141
+ "step": 11000
142
+ },
143
+ {
144
+ "epoch": 0.09,
145
+ "learning_rate": 9.998770068384198e-06,
146
+ "loss": 6.4536,
147
+ "step": 11500
148
+ },
149
+ {
150
+ "epoch": 0.1,
151
+ "learning_rate": 9.99836009117893e-06,
152
+ "loss": 6.443,
153
+ "step": 12000
154
+ },
155
+ {
156
+ "epoch": 0.1,
157
+ "learning_rate": 9.997950113973663e-06,
158
+ "loss": 6.42,
159
+ "step": 12500
160
+ },
161
+ {
162
+ "epoch": 0.11,
163
+ "learning_rate": 9.997540136768397e-06,
164
+ "loss": 6.4115,
165
+ "step": 13000
166
+ },
167
+ {
168
+ "epoch": 0.11,
169
+ "learning_rate": 9.997130159563128e-06,
170
+ "loss": 6.3957,
171
+ "step": 13500
172
+ },
173
+ {
174
+ "epoch": 0.11,
175
+ "learning_rate": 9.996720182357862e-06,
176
+ "loss": 6.386,
177
+ "step": 14000
178
+ },
179
+ {
180
+ "epoch": 0.12,
181
+ "learning_rate": 9.996310205152595e-06,
182
+ "loss": 6.3674,
183
+ "step": 14500
184
+ },
185
+ {
186
+ "epoch": 0.12,
187
+ "learning_rate": 9.995900227947327e-06,
188
+ "loss": 6.3595,
189
+ "step": 15000
190
+ },
191
+ {
192
+ "epoch": 0.13,
193
+ "learning_rate": 9.99549025074206e-06,
194
+ "loss": 6.3477,
195
+ "step": 15500
196
+ },
197
+ {
198
+ "epoch": 0.13,
199
+ "learning_rate": 9.995080273536792e-06,
200
+ "loss": 6.3465,
201
+ "step": 16000
202
+ },
203
+ {
204
+ "epoch": 0.14,
205
+ "learning_rate": 9.994670296331524e-06,
206
+ "loss": 6.3256,
207
+ "step": 16500
208
+ },
209
+ {
210
+ "epoch": 0.14,
211
+ "learning_rate": 9.994260319126257e-06,
212
+ "loss": 6.3203,
213
+ "step": 17000
214
+ },
215
+ {
216
+ "epoch": 0.14,
217
+ "learning_rate": 9.99385034192099e-06,
218
+ "loss": 6.3142,
219
+ "step": 17500
220
+ },
221
+ {
222
+ "epoch": 0.15,
223
+ "learning_rate": 9.993440364715724e-06,
224
+ "loss": 6.2985,
225
+ "step": 18000
226
+ },
227
+ {
228
+ "epoch": 0.15,
229
+ "learning_rate": 9.993030387510454e-06,
230
+ "loss": 6.2923,
231
+ "step": 18500
232
+ },
233
+ {
234
+ "epoch": 0.16,
235
+ "learning_rate": 9.992620410305188e-06,
236
+ "loss": 6.2835,
237
+ "step": 19000
238
+ },
239
+ {
240
+ "epoch": 0.16,
241
+ "learning_rate": 9.992210433099921e-06,
242
+ "loss": 6.2804,
243
+ "step": 19500
244
+ },
245
+ {
246
+ "epoch": 0.16,
247
+ "learning_rate": 9.991800455894653e-06,
248
+ "loss": 6.2699,
249
+ "step": 20000
250
+ },
251
+ {
252
+ "epoch": 0.17,
253
+ "learning_rate": 9.991390478689386e-06,
254
+ "loss": 6.2609,
255
+ "step": 20500
256
+ },
257
+ {
258
+ "epoch": 0.17,
259
+ "learning_rate": 9.990980501484118e-06,
260
+ "loss": 6.2614,
261
+ "step": 21000
262
+ },
263
+ {
264
+ "epoch": 0.18,
265
+ "learning_rate": 9.99057052427885e-06,
266
+ "loss": 6.2472,
267
+ "step": 21500
268
+ },
269
+ {
270
+ "epoch": 0.18,
271
+ "learning_rate": 9.990160547073583e-06,
272
+ "loss": 6.2437,
273
+ "step": 22000
274
+ },
275
+ {
276
+ "epoch": 0.18,
277
+ "learning_rate": 9.989750569868316e-06,
278
+ "loss": 6.2405,
279
+ "step": 22500
280
+ },
281
+ {
282
+ "epoch": 0.19,
283
+ "learning_rate": 9.989340592663048e-06,
284
+ "loss": 6.2412,
285
+ "step": 23000
286
+ },
287
+ {
288
+ "epoch": 0.19,
289
+ "learning_rate": 9.98893061545778e-06,
290
+ "loss": 6.235,
291
+ "step": 23500
292
+ },
293
+ {
294
+ "epoch": 0.2,
295
+ "learning_rate": 9.988520638252513e-06,
296
+ "loss": 6.2297,
297
+ "step": 24000
298
+ },
299
+ {
300
+ "epoch": 0.2,
301
+ "learning_rate": 9.988110661047247e-06,
302
+ "loss": 6.2182,
303
+ "step": 24500
304
+ },
305
+ {
306
+ "epoch": 0.2,
307
+ "learning_rate": 9.987700683841978e-06,
308
+ "loss": 6.2185,
309
+ "step": 25000
310
+ },
311
+ {
312
+ "epoch": 0.21,
313
+ "learning_rate": 9.987290706636712e-06,
314
+ "loss": 6.2121,
315
+ "step": 25500
316
+ },
317
+ {
318
+ "epoch": 0.21,
319
+ "learning_rate": 9.986880729431445e-06,
320
+ "loss": 6.2154,
321
+ "step": 26000
322
+ },
323
+ {
324
+ "epoch": 0.22,
325
+ "learning_rate": 9.986470752226177e-06,
326
+ "loss": 6.2014,
327
+ "step": 26500
328
+ },
329
+ {
330
+ "epoch": 0.22,
331
+ "learning_rate": 9.98606077502091e-06,
332
+ "loss": 6.1937,
333
+ "step": 27000
334
+ },
335
+ {
336
+ "epoch": 0.23,
337
+ "learning_rate": 9.985650797815642e-06,
338
+ "loss": 6.1935,
339
+ "step": 27500
340
+ },
341
+ {
342
+ "epoch": 0.23,
343
+ "learning_rate": 9.985240820610374e-06,
344
+ "loss": 6.1982,
345
+ "step": 28000
346
+ },
347
+ {
348
+ "epoch": 0.23,
349
+ "learning_rate": 9.984830843405109e-06,
350
+ "loss": 6.1961,
351
+ "step": 28500
352
+ },
353
+ {
354
+ "epoch": 0.24,
355
+ "learning_rate": 9.98442086619984e-06,
356
+ "loss": 6.184,
357
+ "step": 29000
358
+ },
359
+ {
360
+ "epoch": 0.24,
361
+ "learning_rate": 9.984010888994574e-06,
362
+ "loss": 6.1825,
363
+ "step": 29500
364
+ },
365
+ {
366
+ "epoch": 0.25,
367
+ "learning_rate": 9.983600911789304e-06,
368
+ "loss": 6.176,
369
+ "step": 30000
370
+ },
371
+ {
372
+ "epoch": 0.25,
373
+ "learning_rate": 9.983190934584038e-06,
374
+ "loss": 6.1768,
375
+ "step": 30500
376
+ },
377
+ {
378
+ "epoch": 0.25,
379
+ "learning_rate": 9.982780957378771e-06,
380
+ "loss": 6.1692,
381
+ "step": 31000
382
+ },
383
+ {
384
+ "epoch": 0.26,
385
+ "learning_rate": 9.982370980173503e-06,
386
+ "loss": 6.1722,
387
+ "step": 31500
388
+ },
389
+ {
390
+ "epoch": 0.26,
391
+ "learning_rate": 9.981961002968236e-06,
392
+ "loss": 6.1705,
393
+ "step": 32000
394
+ },
395
+ {
396
+ "epoch": 0.27,
397
+ "learning_rate": 9.981551025762968e-06,
398
+ "loss": 6.1671,
399
+ "step": 32500
400
+ },
401
+ {
402
+ "epoch": 0.27,
403
+ "learning_rate": 9.9811410485577e-06,
404
+ "loss": 6.1591,
405
+ "step": 33000
406
+ },
407
+ {
408
+ "epoch": 0.27,
409
+ "learning_rate": 9.980731071352435e-06,
410
+ "loss": 6.1511,
411
+ "step": 33500
412
+ },
413
+ {
414
+ "epoch": 0.28,
415
+ "learning_rate": 9.980321094147166e-06,
416
+ "loss": 6.1535,
417
+ "step": 34000
418
+ },
419
+ {
420
+ "epoch": 0.28,
421
+ "learning_rate": 9.979911116941898e-06,
422
+ "loss": 6.154,
423
+ "step": 34500
424
+ },
425
+ {
426
+ "epoch": 0.29,
427
+ "learning_rate": 9.979501139736632e-06,
428
+ "loss": 6.1482,
429
+ "step": 35000
430
+ },
431
+ {
432
+ "epoch": 0.29,
433
+ "learning_rate": 9.979091162531363e-06,
434
+ "loss": 6.1563,
435
+ "step": 35500
436
+ },
437
+ {
438
+ "epoch": 0.29,
439
+ "learning_rate": 9.978681185326097e-06,
440
+ "loss": 6.1401,
441
+ "step": 36000
442
+ },
443
+ {
444
+ "epoch": 0.3,
445
+ "learning_rate": 9.97827120812083e-06,
446
+ "loss": 6.1381,
447
+ "step": 36500
448
+ },
449
+ {
450
+ "epoch": 0.3,
451
+ "learning_rate": 9.977861230915562e-06,
452
+ "loss": 6.1378,
453
+ "step": 37000
454
+ },
455
+ {
456
+ "epoch": 0.31,
457
+ "learning_rate": 9.977451253710295e-06,
458
+ "loss": 6.14,
459
+ "step": 37500
460
+ },
461
+ {
462
+ "epoch": 0.31,
463
+ "learning_rate": 9.977041276505027e-06,
464
+ "loss": 6.1273,
465
+ "step": 38000
466
+ },
467
+ {
468
+ "epoch": 0.32,
469
+ "learning_rate": 9.97663129929976e-06,
470
+ "loss": 6.1371,
471
+ "step": 38500
472
+ },
473
+ {
474
+ "epoch": 0.32,
475
+ "learning_rate": 9.976221322094492e-06,
476
+ "loss": 6.1282,
477
+ "step": 39000
478
+ },
479
+ {
480
+ "epoch": 0.32,
481
+ "learning_rate": 9.975811344889224e-06,
482
+ "loss": 6.1232,
483
+ "step": 39500
484
+ },
485
+ {
486
+ "epoch": 0.33,
487
+ "learning_rate": 9.975401367683959e-06,
488
+ "loss": 6.1257,
489
+ "step": 40000
490
+ },
491
+ {
492
+ "epoch": 0.33,
493
+ "learning_rate": 9.97499139047869e-06,
494
+ "loss": 6.1229,
495
+ "step": 40500
496
+ },
497
+ {
498
+ "epoch": 0.34,
499
+ "learning_rate": 9.974581413273423e-06,
500
+ "loss": 6.123,
501
+ "step": 41000
502
+ },
503
+ {
504
+ "epoch": 0.34,
505
+ "learning_rate": 9.974171436068156e-06,
506
+ "loss": 6.122,
507
+ "step": 41500
508
+ },
509
+ {
510
+ "epoch": 0.34,
511
+ "learning_rate": 9.973761458862888e-06,
512
+ "loss": 6.1213,
513
+ "step": 42000
514
+ },
515
+ {
516
+ "epoch": 0.35,
517
+ "learning_rate": 9.973351481657621e-06,
518
+ "loss": 6.1202,
519
+ "step": 42500
520
+ },
521
+ {
522
+ "epoch": 0.35,
523
+ "learning_rate": 9.972941504452353e-06,
524
+ "loss": 6.1141,
525
+ "step": 43000
526
+ },
527
+ {
528
+ "epoch": 0.36,
529
+ "learning_rate": 9.972531527247086e-06,
530
+ "loss": 6.1135,
531
+ "step": 43500
532
+ },
533
+ {
534
+ "epoch": 0.36,
535
+ "learning_rate": 9.972121550041818e-06,
536
+ "loss": 6.1085,
537
+ "step": 44000
538
+ },
539
+ {
540
+ "epoch": 0.36,
541
+ "learning_rate": 9.97171157283655e-06,
542
+ "loss": 6.1094,
543
+ "step": 44500
544
+ },
545
+ {
546
+ "epoch": 0.37,
547
+ "learning_rate": 9.971301595631285e-06,
548
+ "loss": 6.1085,
549
+ "step": 45000
550
+ },
551
+ {
552
+ "epoch": 0.37,
553
+ "learning_rate": 9.970891618426016e-06,
554
+ "loss": 6.102,
555
+ "step": 45500
556
+ },
557
+ {
558
+ "epoch": 0.38,
559
+ "learning_rate": 9.97048164122075e-06,
560
+ "loss": 6.0967,
561
+ "step": 46000
562
+ },
563
+ {
564
+ "epoch": 0.38,
565
+ "learning_rate": 9.970071664015482e-06,
566
+ "loss": 6.0975,
567
+ "step": 46500
568
+ },
569
+ {
570
+ "epoch": 0.39,
571
+ "learning_rate": 9.969661686810213e-06,
572
+ "loss": 6.0942,
573
+ "step": 47000
574
+ },
575
+ {
576
+ "epoch": 0.39,
577
+ "learning_rate": 9.969251709604947e-06,
578
+ "loss": 6.1022,
579
+ "step": 47500
580
+ },
581
+ {
582
+ "epoch": 0.39,
583
+ "learning_rate": 9.96884173239968e-06,
584
+ "loss": 6.0955,
585
+ "step": 48000
586
+ },
587
+ {
588
+ "epoch": 0.4,
589
+ "learning_rate": 9.968431755194412e-06,
590
+ "loss": 6.0882,
591
+ "step": 48500
592
+ },
593
+ {
594
+ "epoch": 0.4,
595
+ "learning_rate": 9.968021777989145e-06,
596
+ "loss": 6.0863,
597
+ "step": 49000
598
+ },
599
+ {
600
+ "epoch": 0.41,
601
+ "learning_rate": 9.967611800783877e-06,
602
+ "loss": 6.0955,
603
+ "step": 49500
604
+ },
605
+ {
606
+ "epoch": 0.41,
607
+ "learning_rate": 9.96720182357861e-06,
608
+ "loss": 6.0885,
609
+ "step": 50000
610
+ },
611
+ {
612
+ "epoch": 0.41,
613
+ "learning_rate": 9.966791846373342e-06,
614
+ "loss": 6.0898,
615
+ "step": 50500
616
+ },
617
+ {
618
+ "epoch": 0.42,
619
+ "learning_rate": 9.966381869168074e-06,
620
+ "loss": 6.0886,
621
+ "step": 51000
622
+ },
623
+ {
624
+ "epoch": 0.42,
625
+ "learning_rate": 9.965971891962809e-06,
626
+ "loss": 6.0811,
627
+ "step": 51500
628
+ },
629
+ {
630
+ "epoch": 0.43,
631
+ "learning_rate": 9.96556191475754e-06,
632
+ "loss": 6.0758,
633
+ "step": 52000
634
+ },
635
+ {
636
+ "epoch": 0.43,
637
+ "learning_rate": 9.965151937552273e-06,
638
+ "loss": 6.0782,
639
+ "step": 52500
640
+ },
641
+ {
642
+ "epoch": 0.43,
643
+ "learning_rate": 9.964741960347006e-06,
644
+ "loss": 6.0786,
645
+ "step": 53000
646
+ },
647
+ {
648
+ "epoch": 0.44,
649
+ "learning_rate": 9.964331983141738e-06,
650
+ "loss": 6.0729,
651
+ "step": 53500
652
+ },
653
+ {
654
+ "epoch": 0.44,
655
+ "learning_rate": 9.96392200593647e-06,
656
+ "loss": 6.0796,
657
+ "step": 54000
658
+ },
659
+ {
660
+ "epoch": 0.45,
661
+ "learning_rate": 9.963512028731203e-06,
662
+ "loss": 6.0707,
663
+ "step": 54500
664
+ },
665
+ {
666
+ "epoch": 0.45,
667
+ "learning_rate": 9.963102051525936e-06,
668
+ "loss": 6.0731,
669
+ "step": 55000
670
+ },
671
+ {
672
+ "epoch": 0.45,
673
+ "learning_rate": 9.962692074320668e-06,
674
+ "loss": 6.0702,
675
+ "step": 55500
676
+ },
677
+ {
678
+ "epoch": 0.46,
679
+ "learning_rate": 9.9622820971154e-06,
680
+ "loss": 6.073,
681
+ "step": 56000
682
+ },
683
+ {
684
+ "epoch": 0.46,
685
+ "learning_rate": 9.961872119910135e-06,
686
+ "loss": 6.0695,
687
+ "step": 56500
688
+ },
689
+ {
690
+ "epoch": 0.47,
691
+ "learning_rate": 9.961462142704866e-06,
692
+ "loss": 6.0657,
693
+ "step": 57000
694
+ },
695
+ {
696
+ "epoch": 0.47,
697
+ "learning_rate": 9.9610521654996e-06,
698
+ "loss": 6.0668,
699
+ "step": 57500
700
+ },
701
+ {
702
+ "epoch": 0.48,
703
+ "learning_rate": 9.960642188294332e-06,
704
+ "loss": 6.0575,
705
+ "step": 58000
706
+ },
707
+ {
708
+ "epoch": 0.48,
709
+ "learning_rate": 9.960232211089065e-06,
710
+ "loss": 6.0645,
711
+ "step": 58500
712
+ },
713
+ {
714
+ "epoch": 0.48,
715
+ "learning_rate": 9.959822233883797e-06,
716
+ "loss": 6.0605,
717
+ "step": 59000
718
+ },
719
+ {
720
+ "epoch": 0.49,
721
+ "learning_rate": 9.95941225667853e-06,
722
+ "loss": 6.0551,
723
+ "step": 59500
724
+ },
725
+ {
726
+ "epoch": 0.49,
727
+ "learning_rate": 9.959002279473262e-06,
728
+ "loss": 6.0569,
729
+ "step": 60000
730
+ },
731
+ {
732
+ "epoch": 0.5,
733
+ "learning_rate": 9.958592302267995e-06,
734
+ "loss": 6.0596,
735
+ "step": 60500
736
+ },
737
+ {
738
+ "epoch": 0.5,
739
+ "learning_rate": 9.958182325062727e-06,
740
+ "loss": 6.0578,
741
+ "step": 61000
742
+ },
743
+ {
744
+ "epoch": 0.5,
745
+ "learning_rate": 9.95777234785746e-06,
746
+ "loss": 6.0566,
747
+ "step": 61500
748
+ },
749
+ {
750
+ "epoch": 0.51,
751
+ "learning_rate": 9.957362370652192e-06,
752
+ "loss": 6.0571,
753
+ "step": 62000
754
+ },
755
+ {
756
+ "epoch": 0.51,
757
+ "learning_rate": 9.956952393446924e-06,
758
+ "loss": 6.0445,
759
+ "step": 62500
760
+ },
761
+ {
762
+ "epoch": 0.52,
763
+ "learning_rate": 9.956542416241659e-06,
764
+ "loss": 6.0514,
765
+ "step": 63000
766
+ },
767
+ {
768
+ "epoch": 0.52,
769
+ "learning_rate": 9.95613243903639e-06,
770
+ "loss": 6.0461,
771
+ "step": 63500
772
+ },
773
+ {
774
+ "epoch": 0.52,
775
+ "learning_rate": 9.955722461831123e-06,
776
+ "loss": 6.0412,
777
+ "step": 64000
778
+ },
779
+ {
780
+ "epoch": 0.53,
781
+ "learning_rate": 9.955312484625856e-06,
782
+ "loss": 6.0442,
783
+ "step": 64500
784
+ },
785
+ {
786
+ "epoch": 0.53,
787
+ "learning_rate": 9.954902507420588e-06,
788
+ "loss": 6.0488,
789
+ "step": 65000
790
+ },
791
+ {
792
+ "epoch": 0.54,
793
+ "learning_rate": 9.95449253021532e-06,
794
+ "loss": 6.0514,
795
+ "step": 65500
796
+ },
797
+ {
798
+ "epoch": 0.54,
799
+ "learning_rate": 9.954082553010053e-06,
800
+ "loss": 6.0514,
801
+ "step": 66000
802
+ },
803
+ {
804
+ "epoch": 0.54,
805
+ "learning_rate": 9.953672575804786e-06,
806
+ "loss": 6.0415,
807
+ "step": 66500
808
+ },
809
+ {
810
+ "epoch": 0.55,
811
+ "learning_rate": 9.95326259859952e-06,
812
+ "loss": 6.0467,
813
+ "step": 67000
814
+ },
815
+ {
816
+ "epoch": 0.55,
817
+ "learning_rate": 9.95285262139425e-06,
818
+ "loss": 6.0417,
819
+ "step": 67500
820
+ },
821
+ {
822
+ "epoch": 0.56,
823
+ "learning_rate": 9.952442644188985e-06,
824
+ "loss": 6.0354,
825
+ "step": 68000
826
+ },
827
+ {
828
+ "epoch": 0.56,
829
+ "learning_rate": 9.952032666983716e-06,
830
+ "loss": 6.0354,
831
+ "step": 68500
832
+ },
833
+ {
834
+ "epoch": 0.57,
835
+ "learning_rate": 9.95162268977845e-06,
836
+ "loss": 6.0355,
837
+ "step": 69000
838
+ },
839
+ {
840
+ "epoch": 0.57,
841
+ "learning_rate": 9.951212712573182e-06,
842
+ "loss": 6.0326,
843
+ "step": 69500
844
+ },
845
+ {
846
+ "epoch": 0.57,
847
+ "learning_rate": 9.950802735367915e-06,
848
+ "loss": 6.0346,
849
+ "step": 70000
850
+ },
851
+ {
852
+ "epoch": 0.58,
853
+ "learning_rate": 9.950392758162647e-06,
854
+ "loss": 6.0407,
855
+ "step": 70500
856
+ },
857
+ {
858
+ "epoch": 0.58,
859
+ "learning_rate": 9.94998278095738e-06,
860
+ "loss": 6.0378,
861
+ "step": 71000
862
+ },
863
+ {
864
+ "epoch": 0.59,
865
+ "learning_rate": 9.949572803752112e-06,
866
+ "loss": 6.0297,
867
+ "step": 71500
868
+ },
869
+ {
870
+ "epoch": 0.59,
871
+ "learning_rate": 9.949162826546844e-06,
872
+ "loss": 6.0389,
873
+ "step": 72000
874
+ },
875
+ {
876
+ "epoch": 0.59,
877
+ "learning_rate": 9.948752849341577e-06,
878
+ "loss": 6.0293,
879
+ "step": 72500
880
+ },
881
+ {
882
+ "epoch": 0.6,
883
+ "learning_rate": 9.94834287213631e-06,
884
+ "loss": 6.0345,
885
+ "step": 73000
886
+ },
887
+ {
888
+ "epoch": 0.6,
889
+ "learning_rate": 9.947932894931044e-06,
890
+ "loss": 6.0282,
891
+ "step": 73500
892
+ },
893
+ {
894
+ "epoch": 0.61,
895
+ "learning_rate": 9.947522917725774e-06,
896
+ "loss": 6.0257,
897
+ "step": 74000
898
+ },
899
+ {
900
+ "epoch": 0.61,
901
+ "learning_rate": 9.947112940520508e-06,
902
+ "loss": 6.0222,
903
+ "step": 74500
904
+ },
905
+ {
906
+ "epoch": 0.61,
907
+ "learning_rate": 9.946702963315241e-06,
908
+ "loss": 6.0267,
909
+ "step": 75000
910
+ },
911
+ {
912
+ "epoch": 0.62,
913
+ "learning_rate": 9.946292986109973e-06,
914
+ "loss": 6.027,
915
+ "step": 75500
916
+ },
917
+ {
918
+ "epoch": 0.62,
919
+ "learning_rate": 9.945883008904706e-06,
920
+ "loss": 6.0264,
921
+ "step": 76000
922
+ },
923
+ {
924
+ "epoch": 0.63,
925
+ "learning_rate": 9.945473031699438e-06,
926
+ "loss": 6.0267,
927
+ "step": 76500
928
+ },
929
+ {
930
+ "epoch": 0.63,
931
+ "learning_rate": 9.94506305449417e-06,
932
+ "loss": 6.0198,
933
+ "step": 77000
934
+ },
935
+ {
936
+ "epoch": 0.63,
937
+ "learning_rate": 9.944653077288903e-06,
938
+ "loss": 6.0228,
939
+ "step": 77500
940
+ },
941
+ {
942
+ "epoch": 0.64,
943
+ "learning_rate": 9.944243100083636e-06,
944
+ "loss": 6.0162,
945
+ "step": 78000
946
+ },
947
+ {
948
+ "epoch": 0.64,
949
+ "learning_rate": 9.94383312287837e-06,
950
+ "loss": 6.0182,
951
+ "step": 78500
952
+ },
953
+ {
954
+ "epoch": 0.65,
955
+ "learning_rate": 9.9434231456731e-06,
956
+ "loss": 6.0169,
957
+ "step": 79000
958
+ },
959
+ {
960
+ "epoch": 0.65,
961
+ "learning_rate": 9.943013168467835e-06,
962
+ "loss": 6.0175,
963
+ "step": 79500
964
+ },
965
+ {
966
+ "epoch": 0.66,
967
+ "learning_rate": 9.942603191262567e-06,
968
+ "loss": 6.0175,
969
+ "step": 80000
970
+ },
971
+ {
972
+ "epoch": 0.66,
973
+ "learning_rate": 9.9421932140573e-06,
974
+ "loss": 6.0165,
975
+ "step": 80500
976
+ },
977
+ {
978
+ "epoch": 0.66,
979
+ "learning_rate": 9.941783236852032e-06,
980
+ "loss": 6.0144,
981
+ "step": 81000
982
+ },
983
+ {
984
+ "epoch": 0.67,
985
+ "learning_rate": 9.941373259646765e-06,
986
+ "loss": 6.0107,
987
+ "step": 81500
988
+ },
989
+ {
990
+ "epoch": 0.67,
991
+ "learning_rate": 9.940963282441497e-06,
992
+ "loss": 6.0183,
993
+ "step": 82000
994
+ },
995
+ {
996
+ "epoch": 0.68,
997
+ "learning_rate": 9.94055330523623e-06,
998
+ "loss": 6.0146,
999
+ "step": 82500
1000
+ },
1001
+ {
1002
+ "epoch": 0.68,
1003
+ "learning_rate": 9.940143328030962e-06,
1004
+ "loss": 6.0049,
1005
+ "step": 83000
1006
+ },
1007
+ {
1008
+ "epoch": 0.68,
1009
+ "learning_rate": 9.939733350825694e-06,
1010
+ "loss": 6.009,
1011
+ "step": 83500
1012
+ },
1013
+ {
1014
+ "epoch": 0.69,
1015
+ "learning_rate": 9.939323373620427e-06,
1016
+ "loss": 6.009,
1017
+ "step": 84000
1018
+ },
1019
+ {
1020
+ "epoch": 0.69,
1021
+ "learning_rate": 9.93891339641516e-06,
1022
+ "loss": 6.0006,
1023
+ "step": 84500
1024
+ },
1025
+ {
1026
+ "epoch": 0.7,
1027
+ "learning_rate": 9.938503419209894e-06,
1028
+ "loss": 6.0086,
1029
+ "step": 85000
1030
+ },
1031
+ {
1032
+ "epoch": 0.7,
1033
+ "learning_rate": 9.938093442004624e-06,
1034
+ "loss": 6.0052,
1035
+ "step": 85500
1036
+ },
1037
+ {
1038
+ "epoch": 0.7,
1039
+ "learning_rate": 9.937683464799358e-06,
1040
+ "loss": 6.0051,
1041
+ "step": 86000
1042
+ },
1043
+ {
1044
+ "epoch": 0.71,
1045
+ "learning_rate": 9.937273487594091e-06,
1046
+ "loss": 5.9998,
1047
+ "step": 86500
1048
+ },
1049
+ {
1050
+ "epoch": 0.71,
1051
+ "learning_rate": 9.936863510388823e-06,
1052
+ "loss": 6.0042,
1053
+ "step": 87000
1054
+ },
1055
+ {
1056
+ "epoch": 0.72,
1057
+ "learning_rate": 9.936453533183556e-06,
1058
+ "loss": 6.0022,
1059
+ "step": 87500
1060
+ },
1061
+ {
1062
+ "epoch": 0.72,
1063
+ "learning_rate": 9.936043555978288e-06,
1064
+ "loss": 5.9982,
1065
+ "step": 88000
1066
+ },
1067
+ {
1068
+ "epoch": 0.73,
1069
+ "learning_rate": 9.93563357877302e-06,
1070
+ "loss": 6.001,
1071
+ "step": 88500
1072
+ },
1073
+ {
1074
+ "epoch": 0.73,
1075
+ "learning_rate": 9.935223601567753e-06,
1076
+ "loss": 6.006,
1077
+ "step": 89000
1078
+ },
1079
+ {
1080
+ "epoch": 0.73,
1081
+ "learning_rate": 9.934813624362486e-06,
1082
+ "loss": 6.0025,
1083
+ "step": 89500
1084
+ },
1085
+ {
1086
+ "epoch": 0.74,
1087
+ "learning_rate": 9.93440364715722e-06,
1088
+ "loss": 6.0006,
1089
+ "step": 90000
1090
+ },
1091
+ {
1092
+ "epoch": 0.74,
1093
+ "learning_rate": 9.93399366995195e-06,
1094
+ "loss": 6.0028,
1095
+ "step": 90500
1096
+ },
1097
+ {
1098
+ "epoch": 0.75,
1099
+ "learning_rate": 9.933583692746685e-06,
1100
+ "loss": 5.9939,
1101
+ "step": 91000
1102
+ },
1103
+ {
1104
+ "epoch": 0.75,
1105
+ "learning_rate": 9.933173715541417e-06,
1106
+ "loss": 5.996,
1107
+ "step": 91500
1108
+ },
1109
+ {
1110
+ "epoch": 0.75,
1111
+ "learning_rate": 9.93276373833615e-06,
1112
+ "loss": 5.9939,
1113
+ "step": 92000
1114
+ },
1115
+ {
1116
+ "epoch": 0.76,
1117
+ "learning_rate": 9.932353761130882e-06,
1118
+ "loss": 5.9902,
1119
+ "step": 92500
1120
+ },
1121
+ {
1122
+ "epoch": 0.76,
1123
+ "learning_rate": 9.931943783925615e-06,
1124
+ "loss": 5.9998,
1125
+ "step": 93000
1126
+ },
1127
+ {
1128
+ "epoch": 0.77,
1129
+ "learning_rate": 9.931533806720347e-06,
1130
+ "loss": 5.9962,
1131
+ "step": 93500
1132
+ },
1133
+ {
1134
+ "epoch": 0.77,
1135
+ "learning_rate": 9.93112382951508e-06,
1136
+ "loss": 5.99,
1137
+ "step": 94000
1138
+ },
1139
+ {
1140
+ "epoch": 0.77,
1141
+ "learning_rate": 9.930713852309812e-06,
1142
+ "loss": 5.9908,
1143
+ "step": 94500
1144
+ },
1145
+ {
1146
+ "epoch": 0.78,
1147
+ "learning_rate": 9.930303875104546e-06,
1148
+ "loss": 5.9858,
1149
+ "step": 95000
1150
+ },
1151
+ {
1152
+ "epoch": 0.78,
1153
+ "learning_rate": 9.929893897899277e-06,
1154
+ "loss": 5.9929,
1155
+ "step": 95500
1156
+ },
1157
+ {
1158
+ "epoch": 0.79,
1159
+ "learning_rate": 9.92948392069401e-06,
1160
+ "loss": 5.9884,
1161
+ "step": 96000
1162
+ },
1163
+ {
1164
+ "epoch": 0.79,
1165
+ "learning_rate": 9.929073943488743e-06,
1166
+ "loss": 5.9867,
1167
+ "step": 96500
1168
+ },
1169
+ {
1170
+ "epoch": 0.79,
1171
+ "learning_rate": 9.928663966283474e-06,
1172
+ "loss": 5.9906,
1173
+ "step": 97000
1174
+ },
1175
+ {
1176
+ "epoch": 0.8,
1177
+ "learning_rate": 9.928253989078208e-06,
1178
+ "loss": 5.9853,
1179
+ "step": 97500
1180
+ },
1181
+ {
1182
+ "epoch": 0.8,
1183
+ "learning_rate": 9.927844011872941e-06,
1184
+ "loss": 5.9872,
1185
+ "step": 98000
1186
+ },
1187
+ {
1188
+ "epoch": 0.81,
1189
+ "learning_rate": 9.927434034667673e-06,
1190
+ "loss": 5.9921,
1191
+ "step": 98500
1192
+ },
1193
+ {
1194
+ "epoch": 0.81,
1195
+ "learning_rate": 9.927024057462406e-06,
1196
+ "loss": 5.9891,
1197
+ "step": 99000
1198
+ },
1199
+ {
1200
+ "epoch": 0.82,
1201
+ "learning_rate": 9.926614080257138e-06,
1202
+ "loss": 5.9891,
1203
+ "step": 99500
1204
+ },
1205
+ {
1206
+ "epoch": 0.82,
1207
+ "learning_rate": 9.92620410305187e-06,
1208
+ "loss": 5.9899,
1209
+ "step": 100000
1210
+ },
1211
+ {
1212
+ "epoch": 0.82,
1213
+ "learning_rate": 9.925794125846603e-06,
1214
+ "loss": 5.9839,
1215
+ "step": 100500
1216
+ },
1217
+ {
1218
+ "epoch": 0.83,
1219
+ "learning_rate": 9.925384148641336e-06,
1220
+ "loss": 5.9795,
1221
+ "step": 101000
1222
+ },
1223
+ {
1224
+ "epoch": 0.83,
1225
+ "learning_rate": 9.92497417143607e-06,
1226
+ "loss": 5.9819,
1227
+ "step": 101500
1228
+ },
1229
+ {
1230
+ "epoch": 0.84,
1231
+ "learning_rate": 9.9245641942308e-06,
1232
+ "loss": 5.9855,
1233
+ "step": 102000
1234
+ },
1235
+ {
1236
+ "epoch": 0.84,
1237
+ "learning_rate": 9.924154217025535e-06,
1238
+ "loss": 5.9844,
1239
+ "step": 102500
1240
+ },
1241
+ {
1242
+ "epoch": 0.84,
1243
+ "learning_rate": 9.923744239820267e-06,
1244
+ "loss": 5.9821,
1245
+ "step": 103000
1246
+ },
1247
+ {
1248
+ "epoch": 0.85,
1249
+ "learning_rate": 9.923334262615e-06,
1250
+ "loss": 5.9788,
1251
+ "step": 103500
1252
+ },
1253
+ {
1254
+ "epoch": 0.85,
1255
+ "learning_rate": 9.922924285409732e-06,
1256
+ "loss": 5.9848,
1257
+ "step": 104000
1258
+ },
1259
+ {
1260
+ "epoch": 0.86,
1261
+ "learning_rate": 9.922514308204465e-06,
1262
+ "loss": 5.9758,
1263
+ "step": 104500
1264
+ },
1265
+ {
1266
+ "epoch": 0.86,
1267
+ "learning_rate": 9.922104330999197e-06,
1268
+ "loss": 5.9741,
1269
+ "step": 105000
1270
+ },
1271
+ {
1272
+ "epoch": 0.86,
1273
+ "learning_rate": 9.921694353793931e-06,
1274
+ "loss": 5.9756,
1275
+ "step": 105500
1276
+ },
1277
+ {
1278
+ "epoch": 0.87,
1279
+ "learning_rate": 9.921284376588662e-06,
1280
+ "loss": 5.9779,
1281
+ "step": 106000
1282
+ },
1283
+ {
1284
+ "epoch": 0.87,
1285
+ "learning_rate": 9.920874399383396e-06,
1286
+ "loss": 5.9814,
1287
+ "step": 106500
1288
+ },
1289
+ {
1290
+ "epoch": 0.88,
1291
+ "learning_rate": 9.920464422178127e-06,
1292
+ "loss": 5.9843,
1293
+ "step": 107000
1294
+ },
1295
+ {
1296
+ "epoch": 0.88,
1297
+ "learning_rate": 9.920054444972861e-06,
1298
+ "loss": 5.9837,
1299
+ "step": 107500
1300
+ },
1301
+ {
1302
+ "epoch": 0.88,
1303
+ "learning_rate": 9.919644467767593e-06,
1304
+ "loss": 5.9723,
1305
+ "step": 108000
1306
+ },
1307
+ {
1308
+ "epoch": 0.89,
1309
+ "learning_rate": 9.919234490562324e-06,
1310
+ "loss": 5.9738,
1311
+ "step": 108500
1312
+ },
1313
+ {
1314
+ "epoch": 0.89,
1315
+ "learning_rate": 9.918824513357058e-06,
1316
+ "loss": 5.9826,
1317
+ "step": 109000
1318
+ },
1319
+ {
1320
+ "epoch": 0.9,
1321
+ "learning_rate": 9.91841453615179e-06,
1322
+ "loss": 5.9774,
1323
+ "step": 109500
1324
+ },
1325
+ {
1326
+ "epoch": 0.9,
1327
+ "learning_rate": 9.918004558946523e-06,
1328
+ "loss": 5.9758,
1329
+ "step": 110000
1330
+ },
1331
+ {
1332
+ "epoch": 0.91,
1333
+ "learning_rate": 9.917594581741256e-06,
1334
+ "loss": 5.9742,
1335
+ "step": 110500
1336
+ },
1337
+ {
1338
+ "epoch": 0.91,
1339
+ "learning_rate": 9.917184604535988e-06,
1340
+ "loss": 5.9738,
1341
+ "step": 111000
1342
+ },
1343
+ {
1344
+ "epoch": 0.91,
1345
+ "learning_rate": 9.91677462733072e-06,
1346
+ "loss": 5.975,
1347
+ "step": 111500
1348
+ },
1349
+ {
1350
+ "epoch": 0.92,
1351
+ "learning_rate": 9.916364650125455e-06,
1352
+ "loss": 5.9754,
1353
+ "step": 112000
1354
+ },
1355
+ {
1356
+ "epoch": 0.92,
1357
+ "learning_rate": 9.915954672920186e-06,
1358
+ "loss": 5.9696,
1359
+ "step": 112500
1360
+ },
1361
+ {
1362
+ "epoch": 0.93,
1363
+ "learning_rate": 9.91554469571492e-06,
1364
+ "loss": 5.9688,
1365
+ "step": 113000
1366
+ },
1367
+ {
1368
+ "epoch": 0.93,
1369
+ "learning_rate": 9.915134718509652e-06,
1370
+ "loss": 5.9689,
1371
+ "step": 113500
1372
+ },
1373
+ {
1374
+ "epoch": 0.93,
1375
+ "learning_rate": 9.914724741304385e-06,
1376
+ "loss": 5.974,
1377
+ "step": 114000
1378
+ },
1379
+ {
1380
+ "epoch": 0.94,
1381
+ "learning_rate": 9.914314764099117e-06,
1382
+ "loss": 5.9604,
1383
+ "step": 114500
1384
+ },
1385
+ {
1386
+ "epoch": 0.94,
1387
+ "learning_rate": 9.91390478689385e-06,
1388
+ "loss": 5.9666,
1389
+ "step": 115000
1390
+ },
1391
+ {
1392
+ "epoch": 0.95,
1393
+ "learning_rate": 9.913494809688582e-06,
1394
+ "loss": 5.9663,
1395
+ "step": 115500
1396
+ },
1397
+ {
1398
+ "epoch": 0.95,
1399
+ "learning_rate": 9.913084832483315e-06,
1400
+ "loss": 5.9696,
1401
+ "step": 116000
1402
+ },
1403
+ {
1404
+ "epoch": 0.95,
1405
+ "learning_rate": 9.912674855278047e-06,
1406
+ "loss": 5.9652,
1407
+ "step": 116500
1408
+ },
1409
+ {
1410
+ "epoch": 0.96,
1411
+ "learning_rate": 9.912264878072781e-06,
1412
+ "loss": 5.9625,
1413
+ "step": 117000
1414
+ },
1415
+ {
1416
+ "epoch": 0.96,
1417
+ "learning_rate": 9.911854900867512e-06,
1418
+ "loss": 5.9565,
1419
+ "step": 117500
1420
+ },
1421
+ {
1422
+ "epoch": 0.97,
1423
+ "learning_rate": 9.911444923662246e-06,
1424
+ "loss": 5.9561,
1425
+ "step": 118000
1426
+ },
1427
+ {
1428
+ "epoch": 0.97,
1429
+ "learning_rate": 9.911034946456978e-06,
1430
+ "loss": 5.95,
1431
+ "step": 118500
1432
+ },
1433
+ {
1434
+ "epoch": 0.97,
1435
+ "learning_rate": 9.910624969251711e-06,
1436
+ "loss": 5.9468,
1437
+ "step": 119000
1438
+ },
1439
+ {
1440
+ "epoch": 0.98,
1441
+ "learning_rate": 9.910214992046443e-06,
1442
+ "loss": 5.9412,
1443
+ "step": 119500
1444
+ },
1445
+ {
1446
+ "epoch": 0.98,
1447
+ "learning_rate": 9.909805014841176e-06,
1448
+ "loss": 5.9278,
1449
+ "step": 120000
1450
+ },
1451
+ {
1452
+ "epoch": 0.99,
1453
+ "learning_rate": 9.909395037635908e-06,
1454
+ "loss": 5.9177,
1455
+ "step": 120500
1456
+ },
1457
+ {
1458
+ "epoch": 0.99,
1459
+ "learning_rate": 9.90898506043064e-06,
1460
+ "loss": 5.9108,
1461
+ "step": 121000
1462
+ },
1463
+ {
1464
+ "epoch": 1.0,
1465
+ "learning_rate": 9.908575083225373e-06,
1466
+ "loss": 5.8951,
1467
+ "step": 121500
1468
+ },
1469
+ {
1470
+ "epoch": 1.0,
1471
+ "learning_rate": 9.908165106020106e-06,
1472
+ "loss": 5.8845,
1473
+ "step": 122000
1474
+ },
1475
+ {
1476
+ "epoch": 1.0,
1477
+ "learning_rate": 9.907755128814838e-06,
1478
+ "loss": 5.8666,
1479
+ "step": 122500
1480
+ },
1481
+ {
1482
+ "epoch": 1.01,
1483
+ "learning_rate": 9.90734515160957e-06,
1484
+ "loss": 5.8549,
1485
+ "step": 123000
1486
+ },
1487
+ {
1488
+ "epoch": 1.01,
1489
+ "learning_rate": 9.906935174404305e-06,
1490
+ "loss": 5.8484,
1491
+ "step": 123500
1492
+ },
1493
+ {
1494
+ "epoch": 1.02,
1495
+ "learning_rate": 9.906525197199036e-06,
1496
+ "loss": 5.8272,
1497
+ "step": 124000
1498
+ },
1499
+ {
1500
+ "epoch": 1.02,
1501
+ "learning_rate": 9.90611521999377e-06,
1502
+ "loss": 5.8294,
1503
+ "step": 124500
1504
+ },
1505
+ {
1506
+ "epoch": 1.02,
1507
+ "learning_rate": 9.905705242788502e-06,
1508
+ "loss": 5.8177,
1509
+ "step": 125000
1510
+ },
1511
+ {
1512
+ "epoch": 1.03,
1513
+ "learning_rate": 9.905295265583235e-06,
1514
+ "loss": 5.8042,
1515
+ "step": 125500
1516
+ },
1517
+ {
1518
+ "epoch": 1.03,
1519
+ "learning_rate": 9.904885288377967e-06,
1520
+ "loss": 5.7966,
1521
+ "step": 126000
1522
+ },
1523
+ {
1524
+ "epoch": 1.04,
1525
+ "learning_rate": 9.9044753111727e-06,
1526
+ "loss": 5.7856,
1527
+ "step": 126500
1528
+ },
1529
+ {
1530
+ "epoch": 1.04,
1531
+ "learning_rate": 9.904065333967432e-06,
1532
+ "loss": 5.7805,
1533
+ "step": 127000
1534
+ },
1535
+ {
1536
+ "epoch": 1.04,
1537
+ "learning_rate": 9.903655356762164e-06,
1538
+ "loss": 5.765,
1539
+ "step": 127500
1540
+ },
1541
+ {
1542
+ "epoch": 1.05,
1543
+ "learning_rate": 9.903245379556897e-06,
1544
+ "loss": 5.7572,
1545
+ "step": 128000
1546
+ },
1547
+ {
1548
+ "epoch": 1.05,
1549
+ "learning_rate": 9.902835402351631e-06,
1550
+ "loss": 5.7467,
1551
+ "step": 128500
1552
+ },
1553
+ {
1554
+ "epoch": 1.06,
1555
+ "learning_rate": 9.902425425146362e-06,
1556
+ "loss": 5.7438,
1557
+ "step": 129000
1558
+ },
1559
+ {
1560
+ "epoch": 1.06,
1561
+ "learning_rate": 9.902015447941096e-06,
1562
+ "loss": 5.7373,
1563
+ "step": 129500
1564
+ },
1565
+ {
1566
+ "epoch": 1.07,
1567
+ "learning_rate": 9.901605470735828e-06,
1568
+ "loss": 5.7314,
1569
+ "step": 130000
1570
+ },
1571
+ {
1572
+ "epoch": 1.07,
1573
+ "learning_rate": 9.901195493530561e-06,
1574
+ "loss": 5.7186,
1575
+ "step": 130500
1576
+ },
1577
+ {
1578
+ "epoch": 1.07,
1579
+ "learning_rate": 9.900785516325293e-06,
1580
+ "loss": 5.718,
1581
+ "step": 131000
1582
+ },
1583
+ {
1584
+ "epoch": 1.08,
1585
+ "learning_rate": 9.900375539120026e-06,
1586
+ "loss": 5.7112,
1587
+ "step": 131500
1588
+ },
1589
+ {
1590
+ "epoch": 1.08,
1591
+ "learning_rate": 9.899965561914758e-06,
1592
+ "loss": 5.6979,
1593
+ "step": 132000
1594
+ },
1595
+ {
1596
+ "epoch": 1.09,
1597
+ "learning_rate": 9.89955558470949e-06,
1598
+ "loss": 5.6973,
1599
+ "step": 132500
1600
+ },
1601
+ {
1602
+ "epoch": 1.09,
1603
+ "learning_rate": 9.899145607504223e-06,
1604
+ "loss": 5.685,
1605
+ "step": 133000
1606
+ },
1607
+ {
1608
+ "epoch": 1.09,
1609
+ "learning_rate": 9.898735630298956e-06,
1610
+ "loss": 5.6754,
1611
+ "step": 133500
1612
+ },
1613
+ {
1614
+ "epoch": 1.1,
1615
+ "learning_rate": 9.898325653093688e-06,
1616
+ "loss": 5.6818,
1617
+ "step": 134000
1618
+ },
1619
+ {
1620
+ "epoch": 1.1,
1621
+ "learning_rate": 9.89791567588842e-06,
1622
+ "loss": 5.6681,
1623
+ "step": 134500
1624
+ },
1625
+ {
1626
+ "epoch": 1.11,
1627
+ "learning_rate": 9.897505698683155e-06,
1628
+ "loss": 5.662,
1629
+ "step": 135000
1630
+ },
1631
+ {
1632
+ "epoch": 1.11,
1633
+ "learning_rate": 9.897095721477886e-06,
1634
+ "loss": 5.6553,
1635
+ "step": 135500
1636
+ },
1637
+ {
1638
+ "epoch": 1.11,
1639
+ "learning_rate": 9.89668574427262e-06,
1640
+ "loss": 5.6435,
1641
+ "step": 136000
1642
+ },
1643
+ {
1644
+ "epoch": 1.12,
1645
+ "learning_rate": 9.896275767067352e-06,
1646
+ "loss": 5.6417,
1647
+ "step": 136500
1648
+ },
1649
+ {
1650
+ "epoch": 1.12,
1651
+ "learning_rate": 9.895865789862085e-06,
1652
+ "loss": 5.6315,
1653
+ "step": 137000
1654
+ },
1655
+ {
1656
+ "epoch": 1.13,
1657
+ "learning_rate": 9.895455812656817e-06,
1658
+ "loss": 5.6243,
1659
+ "step": 137500
1660
+ },
1661
+ {
1662
+ "epoch": 1.13,
1663
+ "learning_rate": 9.89504583545155e-06,
1664
+ "loss": 5.6162,
1665
+ "step": 138000
1666
+ },
1667
+ {
1668
+ "epoch": 1.13,
1669
+ "learning_rate": 9.894635858246282e-06,
1670
+ "loss": 5.6064,
1671
+ "step": 138500
1672
+ },
1673
+ {
1674
+ "epoch": 1.14,
1675
+ "learning_rate": 9.894225881041014e-06,
1676
+ "loss": 5.6047,
1677
+ "step": 139000
1678
+ },
1679
+ {
1680
+ "epoch": 1.14,
1681
+ "learning_rate": 9.893815903835747e-06,
1682
+ "loss": 5.598,
1683
+ "step": 139500
1684
+ },
1685
+ {
1686
+ "epoch": 1.15,
1687
+ "learning_rate": 9.893405926630481e-06,
1688
+ "loss": 5.5944,
1689
+ "step": 140000
1690
+ },
1691
+ {
1692
+ "epoch": 1.15,
1693
+ "learning_rate": 9.892995949425212e-06,
1694
+ "loss": 5.5791,
1695
+ "step": 140500
1696
+ },
1697
+ {
1698
+ "epoch": 1.16,
1699
+ "learning_rate": 9.892585972219946e-06,
1700
+ "loss": 5.5754,
1701
+ "step": 141000
1702
+ },
1703
+ {
1704
+ "epoch": 1.16,
1705
+ "learning_rate": 9.892175995014678e-06,
1706
+ "loss": 5.5592,
1707
+ "step": 141500
1708
+ },
1709
+ {
1710
+ "epoch": 1.16,
1711
+ "learning_rate": 9.891766017809411e-06,
1712
+ "loss": 5.5599,
1713
+ "step": 142000
1714
+ },
1715
+ {
1716
+ "epoch": 1.17,
1717
+ "learning_rate": 9.891356040604143e-06,
1718
+ "loss": 5.5509,
1719
+ "step": 142500
1720
+ },
1721
+ {
1722
+ "epoch": 1.17,
1723
+ "learning_rate": 9.890946063398876e-06,
1724
+ "loss": 5.5441,
1725
+ "step": 143000
1726
+ },
1727
+ {
1728
+ "epoch": 1.18,
1729
+ "learning_rate": 9.890536086193608e-06,
1730
+ "loss": 5.5397,
1731
+ "step": 143500
1732
+ },
1733
+ {
1734
+ "epoch": 1.18,
1735
+ "learning_rate": 9.890126108988342e-06,
1736
+ "loss": 5.5273,
1737
+ "step": 144000
1738
+ },
1739
+ {
1740
+ "epoch": 1.18,
1741
+ "learning_rate": 9.889716131783073e-06,
1742
+ "loss": 5.5151,
1743
+ "step": 144500
1744
+ },
1745
+ {
1746
+ "epoch": 1.19,
1747
+ "learning_rate": 9.889306154577806e-06,
1748
+ "loss": 5.5035,
1749
+ "step": 145000
1750
+ },
1751
+ {
1752
+ "epoch": 1.19,
1753
+ "learning_rate": 9.888896177372538e-06,
1754
+ "loss": 5.4958,
1755
+ "step": 145500
1756
+ },
1757
+ {
1758
+ "epoch": 1.2,
1759
+ "learning_rate": 9.88848620016727e-06,
1760
+ "loss": 5.4898,
1761
+ "step": 146000
1762
+ },
1763
+ {
1764
+ "epoch": 1.2,
1765
+ "learning_rate": 9.888076222962005e-06,
1766
+ "loss": 5.4751,
1767
+ "step": 146500
1768
+ },
1769
+ {
1770
+ "epoch": 1.2,
1771
+ "learning_rate": 9.887666245756735e-06,
1772
+ "loss": 5.4626,
1773
+ "step": 147000
1774
+ },
1775
+ {
1776
+ "epoch": 1.21,
1777
+ "learning_rate": 9.88725626855147e-06,
1778
+ "loss": 5.4522,
1779
+ "step": 147500
1780
+ },
1781
+ {
1782
+ "epoch": 1.21,
1783
+ "learning_rate": 9.886846291346202e-06,
1784
+ "loss": 5.4701,
1785
+ "step": 148000
1786
+ },
1787
+ {
1788
+ "epoch": 1.22,
1789
+ "learning_rate": 9.886436314140935e-06,
1790
+ "loss": 5.4391,
1791
+ "step": 148500
1792
+ },
1793
+ {
1794
+ "epoch": 1.22,
1795
+ "learning_rate": 9.886026336935667e-06,
1796
+ "loss": 5.4279,
1797
+ "step": 149000
1798
+ },
1799
+ {
1800
+ "epoch": 1.22,
1801
+ "learning_rate": 9.8856163597304e-06,
1802
+ "loss": 5.4183,
1803
+ "step": 149500
1804
+ },
1805
+ {
1806
+ "epoch": 1.23,
1807
+ "learning_rate": 9.885206382525132e-06,
1808
+ "loss": 5.4053,
1809
+ "step": 150000
1810
+ },
1811
+ {
1812
+ "epoch": 1.23,
1813
+ "learning_rate": 9.884796405319866e-06,
1814
+ "loss": 5.4312,
1815
+ "step": 150500
1816
+ },
1817
+ {
1818
+ "epoch": 1.24,
1819
+ "learning_rate": 9.884386428114597e-06,
1820
+ "loss": 5.3896,
1821
+ "step": 151000
1822
+ },
1823
+ {
1824
+ "epoch": 1.24,
1825
+ "learning_rate": 9.883976450909331e-06,
1826
+ "loss": 5.3819,
1827
+ "step": 151500
1828
+ },
1829
+ {
1830
+ "epoch": 1.25,
1831
+ "learning_rate": 9.883566473704063e-06,
1832
+ "loss": 5.4848,
1833
+ "step": 152000
1834
+ },
1835
+ {
1836
+ "epoch": 1.25,
1837
+ "learning_rate": 9.883156496498796e-06,
1838
+ "loss": 5.4951,
1839
+ "step": 152500
1840
+ },
1841
+ {
1842
+ "epoch": 1.25,
1843
+ "learning_rate": 9.882746519293528e-06,
1844
+ "loss": 5.459,
1845
+ "step": 153000
1846
+ },
1847
+ {
1848
+ "epoch": 1.26,
1849
+ "learning_rate": 9.882336542088261e-06,
1850
+ "loss": 5.4116,
1851
+ "step": 153500
1852
+ },
1853
+ {
1854
+ "epoch": 1.26,
1855
+ "learning_rate": 9.881926564882993e-06,
1856
+ "loss": 5.3916,
1857
+ "step": 154000
1858
+ },
1859
+ {
1860
+ "epoch": 1.27,
1861
+ "learning_rate": 9.881516587677726e-06,
1862
+ "loss": 5.3717,
1863
+ "step": 154500
1864
+ },
1865
+ {
1866
+ "epoch": 1.27,
1867
+ "learning_rate": 9.881106610472458e-06,
1868
+ "loss": 5.3408,
1869
+ "step": 155000
1870
+ },
1871
+ {
1872
+ "epoch": 1.27,
1873
+ "learning_rate": 9.880696633267192e-06,
1874
+ "loss": 5.3338,
1875
+ "step": 155500
1876
+ },
1877
+ {
1878
+ "epoch": 1.28,
1879
+ "learning_rate": 9.880286656061923e-06,
1880
+ "loss": 5.3153,
1881
+ "step": 156000
1882
+ },
1883
+ {
1884
+ "epoch": 1.28,
1885
+ "learning_rate": 9.879876678856657e-06,
1886
+ "loss": 5.2978,
1887
+ "step": 156500
1888
+ },
1889
+ {
1890
+ "epoch": 1.29,
1891
+ "learning_rate": 9.87946670165139e-06,
1892
+ "loss": 5.2771,
1893
+ "step": 157000
1894
+ },
1895
+ {
1896
+ "epoch": 1.29,
1897
+ "learning_rate": 9.87905672444612e-06,
1898
+ "loss": 5.2582,
1899
+ "step": 157500
1900
+ },
1901
+ {
1902
+ "epoch": 1.29,
1903
+ "learning_rate": 9.878646747240855e-06,
1904
+ "loss": 5.2367,
1905
+ "step": 158000
1906
+ },
1907
+ {
1908
+ "epoch": 1.3,
1909
+ "learning_rate": 9.878236770035587e-06,
1910
+ "loss": 5.2109,
1911
+ "step": 158500
1912
+ },
1913
+ {
1914
+ "epoch": 1.3,
1915
+ "learning_rate": 9.87782679283032e-06,
1916
+ "loss": 5.2005,
1917
+ "step": 159000
1918
+ },
1919
+ {
1920
+ "epoch": 1.31,
1921
+ "learning_rate": 9.877416815625052e-06,
1922
+ "loss": 5.1799,
1923
+ "step": 159500
1924
+ },
1925
+ {
1926
+ "epoch": 1.31,
1927
+ "learning_rate": 9.877006838419785e-06,
1928
+ "loss": 5.1683,
1929
+ "step": 160000
1930
+ },
1931
+ {
1932
+ "epoch": 1.31,
1933
+ "learning_rate": 9.876596861214517e-06,
1934
+ "loss": 5.1326,
1935
+ "step": 160500
1936
+ },
1937
+ {
1938
+ "epoch": 1.32,
1939
+ "learning_rate": 9.87618688400925e-06,
1940
+ "loss": 5.1092,
1941
+ "step": 161000
1942
+ },
1943
+ {
1944
+ "epoch": 1.32,
1945
+ "learning_rate": 9.875776906803982e-06,
1946
+ "loss": 5.0876,
1947
+ "step": 161500
1948
+ },
1949
+ {
1950
+ "epoch": 1.33,
1951
+ "learning_rate": 9.875366929598716e-06,
1952
+ "loss": 5.0742,
1953
+ "step": 162000
1954
+ },
1955
+ {
1956
+ "epoch": 1.33,
1957
+ "learning_rate": 9.874956952393447e-06,
1958
+ "loss": 5.0547,
1959
+ "step": 162500
1960
+ },
1961
+ {
1962
+ "epoch": 1.34,
1963
+ "learning_rate": 9.874546975188181e-06,
1964
+ "loss": 5.0256,
1965
+ "step": 163000
1966
+ },
1967
+ {
1968
+ "epoch": 1.34,
1969
+ "learning_rate": 9.874136997982913e-06,
1970
+ "loss": 5.0043,
1971
+ "step": 163500
1972
+ },
1973
+ {
1974
+ "epoch": 1.34,
1975
+ "learning_rate": 9.873727020777646e-06,
1976
+ "loss": 4.9879,
1977
+ "step": 164000
1978
+ },
1979
+ {
1980
+ "epoch": 1.35,
1981
+ "learning_rate": 9.873317043572378e-06,
1982
+ "loss": 4.9613,
1983
+ "step": 164500
1984
+ },
1985
+ {
1986
+ "epoch": 1.35,
1987
+ "learning_rate": 9.87290706636711e-06,
1988
+ "loss": 4.9432,
1989
+ "step": 165000
1990
+ },
1991
+ {
1992
+ "epoch": 1.36,
1993
+ "learning_rate": 9.872497089161843e-06,
1994
+ "loss": 4.9568,
1995
+ "step": 165500
1996
+ },
1997
+ {
1998
+ "epoch": 1.36,
1999
+ "learning_rate": 9.872087111956576e-06,
2000
+ "loss": 4.9095,
2001
+ "step": 166000
2002
+ },
2003
+ {
2004
+ "epoch": 1.36,
2005
+ "learning_rate": 9.871677134751308e-06,
2006
+ "loss": 4.8854,
2007
+ "step": 166500
2008
+ },
2009
+ {
2010
+ "epoch": 1.37,
2011
+ "learning_rate": 9.871267157546042e-06,
2012
+ "loss": 4.8734,
2013
+ "step": 167000
2014
+ },
2015
+ {
2016
+ "epoch": 1.37,
2017
+ "learning_rate": 9.870857180340773e-06,
2018
+ "loss": 4.8439,
2019
+ "step": 167500
2020
+ },
2021
+ {
2022
+ "epoch": 1.38,
2023
+ "learning_rate": 9.870447203135507e-06,
2024
+ "loss": 4.8283,
2025
+ "step": 168000
2026
+ },
2027
+ {
2028
+ "epoch": 1.38,
2029
+ "learning_rate": 9.87003722593024e-06,
2030
+ "loss": 4.7974,
2031
+ "step": 168500
2032
+ },
2033
+ {
2034
+ "epoch": 1.38,
2035
+ "learning_rate": 9.869627248724972e-06,
2036
+ "loss": 4.7818,
2037
+ "step": 169000
2038
+ },
2039
+ {
2040
+ "epoch": 1.39,
2041
+ "learning_rate": 9.869217271519705e-06,
2042
+ "loss": 4.7894,
2043
+ "step": 169500
2044
+ },
2045
+ {
2046
+ "epoch": 1.39,
2047
+ "learning_rate": 9.868807294314437e-06,
2048
+ "loss": 4.7617,
2049
+ "step": 170000
2050
+ },
2051
+ {
2052
+ "epoch": 1.4,
2053
+ "learning_rate": 9.86839731710917e-06,
2054
+ "loss": 4.7221,
2055
+ "step": 170500
2056
+ },
2057
+ {
2058
+ "epoch": 1.4,
2059
+ "learning_rate": 9.867987339903902e-06,
2060
+ "loss": 4.7078,
2061
+ "step": 171000
2062
+ },
2063
+ {
2064
+ "epoch": 1.41,
2065
+ "learning_rate": 9.867577362698634e-06,
2066
+ "loss": 4.6952,
2067
+ "step": 171500
2068
+ },
2069
+ {
2070
+ "epoch": 1.41,
2071
+ "learning_rate": 9.867167385493367e-06,
2072
+ "loss": 4.6771,
2073
+ "step": 172000
2074
+ },
2075
+ {
2076
+ "epoch": 1.41,
2077
+ "learning_rate": 9.8667574082881e-06,
2078
+ "loss": 4.6439,
2079
+ "step": 172500
2080
+ },
2081
+ {
2082
+ "epoch": 1.42,
2083
+ "learning_rate": 9.866347431082832e-06,
2084
+ "loss": 4.6252,
2085
+ "step": 173000
2086
+ },
2087
+ {
2088
+ "epoch": 1.42,
2089
+ "learning_rate": 9.865937453877566e-06,
2090
+ "loss": 4.6202,
2091
+ "step": 173500
2092
+ },
2093
+ {
2094
+ "epoch": 1.43,
2095
+ "learning_rate": 9.865527476672297e-06,
2096
+ "loss": 4.5785,
2097
+ "step": 174000
2098
+ },
2099
+ {
2100
+ "epoch": 1.43,
2101
+ "learning_rate": 9.865117499467031e-06,
2102
+ "loss": 4.5526,
2103
+ "step": 174500
2104
+ },
2105
+ {
2106
+ "epoch": 1.43,
2107
+ "learning_rate": 9.864707522261763e-06,
2108
+ "loss": 4.5272,
2109
+ "step": 175000
2110
+ },
2111
+ {
2112
+ "epoch": 1.44,
2113
+ "learning_rate": 9.864297545056496e-06,
2114
+ "loss": 4.5108,
2115
+ "step": 175500
2116
+ },
2117
+ {
2118
+ "epoch": 1.44,
2119
+ "learning_rate": 9.863887567851228e-06,
2120
+ "loss": 4.4886,
2121
+ "step": 176000
2122
+ },
2123
+ {
2124
+ "epoch": 1.45,
2125
+ "learning_rate": 9.86347759064596e-06,
2126
+ "loss": 4.457,
2127
+ "step": 176500
2128
+ },
2129
+ {
2130
+ "epoch": 1.45,
2131
+ "learning_rate": 9.863067613440693e-06,
2132
+ "loss": 4.4399,
2133
+ "step": 177000
2134
+ },
2135
+ {
2136
+ "epoch": 1.45,
2137
+ "learning_rate": 9.862657636235426e-06,
2138
+ "loss": 4.4196,
2139
+ "step": 177500
2140
+ },
2141
+ {
2142
+ "epoch": 1.46,
2143
+ "learning_rate": 9.862247659030158e-06,
2144
+ "loss": 4.3904,
2145
+ "step": 178000
2146
+ },
2147
+ {
2148
+ "epoch": 1.46,
2149
+ "learning_rate": 9.861837681824892e-06,
2150
+ "loss": 4.3747,
2151
+ "step": 178500
2152
+ },
2153
+ {
2154
+ "epoch": 1.47,
2155
+ "learning_rate": 9.861427704619623e-06,
2156
+ "loss": 4.3499,
2157
+ "step": 179000
2158
+ },
2159
+ {
2160
+ "epoch": 1.47,
2161
+ "learning_rate": 9.861017727414357e-06,
2162
+ "loss": 4.3159,
2163
+ "step": 179500
2164
+ },
2165
+ {
2166
+ "epoch": 1.47,
2167
+ "learning_rate": 9.86060775020909e-06,
2168
+ "loss": 4.2988,
2169
+ "step": 180000
2170
+ },
2171
+ {
2172
+ "epoch": 1.48,
2173
+ "learning_rate": 9.860197773003822e-06,
2174
+ "loss": 4.2792,
2175
+ "step": 180500
2176
+ },
2177
+ {
2178
+ "epoch": 1.48,
2179
+ "learning_rate": 9.859787795798555e-06,
2180
+ "loss": 4.259,
2181
+ "step": 181000
2182
+ },
2183
+ {
2184
+ "epoch": 1.49,
2185
+ "learning_rate": 9.859377818593287e-06,
2186
+ "loss": 4.2489,
2187
+ "step": 181500
2188
+ },
2189
+ {
2190
+ "epoch": 1.49,
2191
+ "learning_rate": 9.85896784138802e-06,
2192
+ "loss": 4.2267,
2193
+ "step": 182000
2194
+ },
2195
+ {
2196
+ "epoch": 1.5,
2197
+ "learning_rate": 9.858557864182752e-06,
2198
+ "loss": 4.1976,
2199
+ "step": 182500
2200
+ },
2201
+ {
2202
+ "epoch": 1.5,
2203
+ "learning_rate": 9.858147886977484e-06,
2204
+ "loss": 4.1874,
2205
+ "step": 183000
2206
+ },
2207
+ {
2208
+ "epoch": 1.5,
2209
+ "learning_rate": 9.857737909772217e-06,
2210
+ "loss": 4.191,
2211
+ "step": 183500
2212
+ },
2213
+ {
2214
+ "epoch": 1.51,
2215
+ "learning_rate": 9.85732793256695e-06,
2216
+ "loss": 4.155,
2217
+ "step": 184000
2218
+ },
2219
+ {
2220
+ "epoch": 1.51,
2221
+ "learning_rate": 9.856917955361682e-06,
2222
+ "loss": 4.1255,
2223
+ "step": 184500
2224
+ },
2225
+ {
2226
+ "epoch": 1.52,
2227
+ "learning_rate": 9.856507978156416e-06,
2228
+ "loss": 4.1147,
2229
+ "step": 185000
2230
+ },
2231
+ {
2232
+ "epoch": 1.52,
2233
+ "learning_rate": 9.856098000951147e-06,
2234
+ "loss": 4.0869,
2235
+ "step": 185500
2236
+ },
2237
+ {
2238
+ "epoch": 1.52,
2239
+ "learning_rate": 9.855688023745881e-06,
2240
+ "loss": 4.0806,
2241
+ "step": 186000
2242
+ },
2243
+ {
2244
+ "epoch": 1.53,
2245
+ "learning_rate": 9.855278046540613e-06,
2246
+ "loss": 4.0635,
2247
+ "step": 186500
2248
+ },
2249
+ {
2250
+ "epoch": 1.53,
2251
+ "learning_rate": 9.854868069335346e-06,
2252
+ "loss": 4.0525,
2253
+ "step": 187000
2254
+ },
2255
+ {
2256
+ "epoch": 1.54,
2257
+ "learning_rate": 9.854458092130078e-06,
2258
+ "loss": 4.0286,
2259
+ "step": 187500
2260
+ },
2261
+ {
2262
+ "epoch": 1.54,
2263
+ "learning_rate": 9.85404811492481e-06,
2264
+ "loss": 4.0121,
2265
+ "step": 188000
2266
+ },
2267
+ {
2268
+ "epoch": 1.54,
2269
+ "learning_rate": 9.853638137719543e-06,
2270
+ "loss": 4.0062,
2271
+ "step": 188500
2272
+ },
2273
+ {
2274
+ "epoch": 1.55,
2275
+ "learning_rate": 9.853228160514277e-06,
2276
+ "loss": 3.9757,
2277
+ "step": 189000
2278
+ },
2279
+ {
2280
+ "epoch": 1.55,
2281
+ "learning_rate": 9.852818183309008e-06,
2282
+ "loss": 3.9754,
2283
+ "step": 189500
2284
+ },
2285
+ {
2286
+ "epoch": 1.56,
2287
+ "learning_rate": 9.852408206103742e-06,
2288
+ "loss": 3.9647,
2289
+ "step": 190000
2290
+ },
2291
+ {
2292
+ "epoch": 1.56,
2293
+ "learning_rate": 9.851998228898475e-06,
2294
+ "loss": 3.9375,
2295
+ "step": 190500
2296
+ },
2297
+ {
2298
+ "epoch": 1.56,
2299
+ "learning_rate": 9.851588251693207e-06,
2300
+ "loss": 4.0285,
2301
+ "step": 191000
2302
+ },
2303
+ {
2304
+ "epoch": 1.57,
2305
+ "learning_rate": 9.85117827448794e-06,
2306
+ "loss": 3.9618,
2307
+ "step": 191500
2308
+ },
2309
+ {
2310
+ "epoch": 1.57,
2311
+ "learning_rate": 9.850768297282672e-06,
2312
+ "loss": 3.9384,
2313
+ "step": 192000
2314
+ },
2315
+ {
2316
+ "epoch": 1.58,
2317
+ "learning_rate": 9.850358320077405e-06,
2318
+ "loss": 3.8983,
2319
+ "step": 192500
2320
+ },
2321
+ {
2322
+ "epoch": 1.58,
2323
+ "learning_rate": 9.849948342872137e-06,
2324
+ "loss": 3.8736,
2325
+ "step": 193000
2326
+ },
2327
+ {
2328
+ "epoch": 1.59,
2329
+ "learning_rate": 9.84953836566687e-06,
2330
+ "loss": 3.8536,
2331
+ "step": 193500
2332
+ },
2333
+ {
2334
+ "epoch": 1.59,
2335
+ "learning_rate": 9.849128388461602e-06,
2336
+ "loss": 3.8413,
2337
+ "step": 194000
2338
+ },
2339
+ {
2340
+ "epoch": 1.59,
2341
+ "learning_rate": 9.848718411256334e-06,
2342
+ "loss": 3.8269,
2343
+ "step": 194500
2344
+ },
2345
+ {
2346
+ "epoch": 1.6,
2347
+ "learning_rate": 9.848308434051067e-06,
2348
+ "loss": 3.8253,
2349
+ "step": 195000
2350
+ },
2351
+ {
2352
+ "epoch": 1.6,
2353
+ "learning_rate": 9.847898456845801e-06,
2354
+ "loss": 3.7981,
2355
+ "step": 195500
2356
+ },
2357
+ {
2358
+ "epoch": 1.61,
2359
+ "learning_rate": 9.847488479640532e-06,
2360
+ "loss": 3.7768,
2361
+ "step": 196000
2362
+ },
2363
+ {
2364
+ "epoch": 1.61,
2365
+ "learning_rate": 9.847078502435266e-06,
2366
+ "loss": 3.7624,
2367
+ "step": 196500
2368
+ },
2369
+ {
2370
+ "epoch": 1.61,
2371
+ "learning_rate": 9.846668525229998e-06,
2372
+ "loss": 3.7615,
2373
+ "step": 197000
2374
+ },
2375
+ {
2376
+ "epoch": 1.62,
2377
+ "learning_rate": 9.846258548024731e-06,
2378
+ "loss": 3.7521,
2379
+ "step": 197500
2380
+ },
2381
+ {
2382
+ "epoch": 1.62,
2383
+ "learning_rate": 9.845848570819463e-06,
2384
+ "loss": 3.7378,
2385
+ "step": 198000
2386
+ },
2387
+ {
2388
+ "epoch": 1.63,
2389
+ "learning_rate": 9.845438593614196e-06,
2390
+ "loss": 3.7196,
2391
+ "step": 198500
2392
+ },
2393
+ {
2394
+ "epoch": 1.63,
2395
+ "learning_rate": 9.845028616408928e-06,
2396
+ "loss": 3.6944,
2397
+ "step": 199000
2398
+ },
2399
+ {
2400
+ "epoch": 1.63,
2401
+ "learning_rate": 9.84461863920366e-06,
2402
+ "loss": 3.6922,
2403
+ "step": 199500
2404
+ },
2405
+ {
2406
+ "epoch": 1.64,
2407
+ "learning_rate": 9.844208661998393e-06,
2408
+ "loss": 3.682,
2409
+ "step": 200000
2410
+ },
2411
+ {
2412
+ "epoch": 1.64,
2413
+ "learning_rate": 9.843798684793127e-06,
2414
+ "loss": 3.6819,
2415
+ "step": 200500
2416
+ },
2417
+ {
2418
+ "epoch": 1.65,
2419
+ "learning_rate": 9.843388707587858e-06,
2420
+ "loss": 3.6617,
2421
+ "step": 201000
2422
+ },
2423
+ {
2424
+ "epoch": 1.65,
2425
+ "learning_rate": 9.842978730382592e-06,
2426
+ "loss": 3.6491,
2427
+ "step": 201500
2428
+ },
2429
+ {
2430
+ "epoch": 1.65,
2431
+ "learning_rate": 9.842568753177325e-06,
2432
+ "loss": 3.6424,
2433
+ "step": 202000
2434
+ },
2435
+ {
2436
+ "epoch": 1.66,
2437
+ "learning_rate": 9.842158775972057e-06,
2438
+ "loss": 3.6295,
2439
+ "step": 202500
2440
+ },
2441
+ {
2442
+ "epoch": 1.66,
2443
+ "learning_rate": 9.84174879876679e-06,
2444
+ "loss": 3.6155,
2445
+ "step": 203000
2446
+ },
2447
+ {
2448
+ "epoch": 1.67,
2449
+ "learning_rate": 9.841338821561522e-06,
2450
+ "loss": 3.6067,
2451
+ "step": 203500
2452
+ },
2453
+ {
2454
+ "epoch": 1.67,
2455
+ "learning_rate": 9.840928844356255e-06,
2456
+ "loss": 3.6006,
2457
+ "step": 204000
2458
+ },
2459
+ {
2460
+ "epoch": 1.68,
2461
+ "learning_rate": 9.840518867150987e-06,
2462
+ "loss": 3.583,
2463
+ "step": 204500
2464
+ },
2465
+ {
2466
+ "epoch": 1.68,
2467
+ "learning_rate": 9.84010888994572e-06,
2468
+ "loss": 3.5762,
2469
+ "step": 205000
2470
+ },
2471
+ {
2472
+ "epoch": 1.68,
2473
+ "learning_rate": 9.839698912740454e-06,
2474
+ "loss": 3.5693,
2475
+ "step": 205500
2476
+ },
2477
+ {
2478
+ "epoch": 1.69,
2479
+ "learning_rate": 9.839288935535184e-06,
2480
+ "loss": 3.5653,
2481
+ "step": 206000
2482
+ },
2483
+ {
2484
+ "epoch": 1.69,
2485
+ "learning_rate": 9.838878958329917e-06,
2486
+ "loss": 3.5646,
2487
+ "step": 206500
2488
+ },
2489
+ {
2490
+ "epoch": 1.7,
2491
+ "learning_rate": 9.838468981124651e-06,
2492
+ "loss": 3.554,
2493
+ "step": 207000
2494
+ },
2495
+ {
2496
+ "epoch": 1.7,
2497
+ "learning_rate": 9.838059003919382e-06,
2498
+ "loss": 3.5415,
2499
+ "step": 207500
2500
+ },
2501
+ {
2502
+ "epoch": 1.7,
2503
+ "learning_rate": 9.837649026714116e-06,
2504
+ "loss": 3.5334,
2505
+ "step": 208000
2506
+ },
2507
+ {
2508
+ "epoch": 1.71,
2509
+ "learning_rate": 9.837239049508848e-06,
2510
+ "loss": 3.5344,
2511
+ "step": 208500
2512
+ },
2513
+ {
2514
+ "epoch": 1.71,
2515
+ "learning_rate": 9.83682907230358e-06,
2516
+ "loss": 3.5127,
2517
+ "step": 209000
2518
+ },
2519
+ {
2520
+ "epoch": 1.72,
2521
+ "learning_rate": 9.836419095098313e-06,
2522
+ "loss": 3.5073,
2523
+ "step": 209500
2524
+ },
2525
+ {
2526
+ "epoch": 1.72,
2527
+ "learning_rate": 9.836009117893046e-06,
2528
+ "loss": 3.5022,
2529
+ "step": 210000
2530
+ },
2531
+ {
2532
+ "epoch": 1.72,
2533
+ "learning_rate": 9.835599140687778e-06,
2534
+ "loss": 3.4953,
2535
+ "step": 210500
2536
+ },
2537
+ {
2538
+ "epoch": 1.73,
2539
+ "learning_rate": 9.83518916348251e-06,
2540
+ "loss": 3.4879,
2541
+ "step": 211000
2542
+ },
2543
+ {
2544
+ "epoch": 1.73,
2545
+ "learning_rate": 9.834779186277243e-06,
2546
+ "loss": 3.4836,
2547
+ "step": 211500
2548
+ },
2549
+ {
2550
+ "epoch": 1.74,
2551
+ "learning_rate": 9.834369209071977e-06,
2552
+ "loss": 3.486,
2553
+ "step": 212000
2554
+ },
2555
+ {
2556
+ "epoch": 1.74,
2557
+ "learning_rate": 9.833959231866708e-06,
2558
+ "loss": 3.4859,
2559
+ "step": 212500
2560
+ },
2561
+ {
2562
+ "epoch": 1.75,
2563
+ "learning_rate": 9.833549254661442e-06,
2564
+ "loss": 3.4669,
2565
+ "step": 213000
2566
+ },
2567
+ {
2568
+ "epoch": 1.75,
2569
+ "learning_rate": 9.833139277456175e-06,
2570
+ "loss": 3.4699,
2571
+ "step": 213500
2572
+ },
2573
+ {
2574
+ "epoch": 1.75,
2575
+ "learning_rate": 9.832729300250907e-06,
2576
+ "loss": 3.4552,
2577
+ "step": 214000
2578
+ },
2579
+ {
2580
+ "epoch": 1.76,
2581
+ "learning_rate": 9.83231932304564e-06,
2582
+ "loss": 3.4458,
2583
+ "step": 214500
2584
+ },
2585
+ {
2586
+ "epoch": 1.76,
2587
+ "learning_rate": 9.831909345840372e-06,
2588
+ "loss": 3.4439,
2589
+ "step": 215000
2590
+ },
2591
+ {
2592
+ "epoch": 1.77,
2593
+ "learning_rate": 9.831499368635105e-06,
2594
+ "loss": 3.4365,
2595
+ "step": 215500
2596
+ },
2597
+ {
2598
+ "epoch": 1.77,
2599
+ "learning_rate": 9.831089391429837e-06,
2600
+ "loss": 3.421,
2601
+ "step": 216000
2602
+ },
2603
+ {
2604
+ "epoch": 1.77,
2605
+ "learning_rate": 9.83067941422457e-06,
2606
+ "loss": 3.4254,
2607
+ "step": 216500
2608
+ },
2609
+ {
2610
+ "epoch": 1.78,
2611
+ "learning_rate": 9.830269437019304e-06,
2612
+ "loss": 3.4283,
2613
+ "step": 217000
2614
+ },
2615
+ {
2616
+ "epoch": 1.78,
2617
+ "learning_rate": 9.829859459814034e-06,
2618
+ "loss": 3.4104,
2619
+ "step": 217500
2620
+ },
2621
+ {
2622
+ "epoch": 1.79,
2623
+ "learning_rate": 9.829449482608769e-06,
2624
+ "loss": 3.4117,
2625
+ "step": 218000
2626
+ },
2627
+ {
2628
+ "epoch": 1.79,
2629
+ "learning_rate": 9.829039505403501e-06,
2630
+ "loss": 3.4137,
2631
+ "step": 218500
2632
+ },
2633
+ {
2634
+ "epoch": 1.79,
2635
+ "learning_rate": 9.828629528198232e-06,
2636
+ "loss": 3.4104,
2637
+ "step": 219000
2638
+ },
2639
+ {
2640
+ "epoch": 1.8,
2641
+ "learning_rate": 9.828219550992966e-06,
2642
+ "loss": 3.4028,
2643
+ "step": 219500
2644
+ },
2645
+ {
2646
+ "epoch": 1.8,
2647
+ "learning_rate": 9.827809573787698e-06,
2648
+ "loss": 3.3965,
2649
+ "step": 220000
2650
+ },
2651
+ {
2652
+ "epoch": 1.81,
2653
+ "learning_rate": 9.82739959658243e-06,
2654
+ "loss": 3.398,
2655
+ "step": 220500
2656
+ },
2657
+ {
2658
+ "epoch": 1.81,
2659
+ "learning_rate": 9.826989619377163e-06,
2660
+ "loss": 3.3841,
2661
+ "step": 221000
2662
+ },
2663
+ {
2664
+ "epoch": 1.81,
2665
+ "learning_rate": 9.826579642171896e-06,
2666
+ "loss": 3.3684,
2667
+ "step": 221500
2668
+ },
2669
+ {
2670
+ "epoch": 1.82,
2671
+ "learning_rate": 9.826169664966628e-06,
2672
+ "loss": 3.3729,
2673
+ "step": 222000
2674
+ },
2675
+ {
2676
+ "epoch": 1.82,
2677
+ "learning_rate": 9.82575968776136e-06,
2678
+ "loss": 3.3701,
2679
+ "step": 222500
2680
+ },
2681
+ {
2682
+ "epoch": 1.83,
2683
+ "learning_rate": 9.825349710556093e-06,
2684
+ "loss": 3.3599,
2685
+ "step": 223000
2686
+ },
2687
+ {
2688
+ "epoch": 1.83,
2689
+ "learning_rate": 9.824939733350827e-06,
2690
+ "loss": 3.3518,
2691
+ "step": 223500
2692
+ },
2693
+ {
2694
+ "epoch": 1.84,
2695
+ "learning_rate": 9.824529756145558e-06,
2696
+ "loss": 3.3591,
2697
+ "step": 224000
2698
+ },
2699
+ {
2700
+ "epoch": 1.84,
2701
+ "learning_rate": 9.824119778940292e-06,
2702
+ "loss": 3.3534,
2703
+ "step": 224500
2704
+ },
2705
+ {
2706
+ "epoch": 1.84,
2707
+ "learning_rate": 9.823709801735025e-06,
2708
+ "loss": 3.3433,
2709
+ "step": 225000
2710
+ },
2711
+ {
2712
+ "epoch": 1.85,
2713
+ "learning_rate": 9.823299824529757e-06,
2714
+ "loss": 3.3323,
2715
+ "step": 225500
2716
+ },
2717
+ {
2718
+ "epoch": 1.85,
2719
+ "learning_rate": 9.82288984732449e-06,
2720
+ "loss": 3.3267,
2721
+ "step": 226000
2722
+ },
2723
+ {
2724
+ "epoch": 1.86,
2725
+ "learning_rate": 9.822479870119222e-06,
2726
+ "loss": 3.3217,
2727
+ "step": 226500
2728
+ },
2729
+ {
2730
+ "epoch": 1.86,
2731
+ "learning_rate": 9.822069892913954e-06,
2732
+ "loss": 3.3107,
2733
+ "step": 227000
2734
+ },
2735
+ {
2736
+ "epoch": 1.86,
2737
+ "learning_rate": 9.821659915708689e-06,
2738
+ "loss": 3.3058,
2739
+ "step": 227500
2740
+ },
2741
+ {
2742
+ "epoch": 1.87,
2743
+ "learning_rate": 9.82124993850342e-06,
2744
+ "loss": 3.3154,
2745
+ "step": 228000
2746
+ },
2747
+ {
2748
+ "epoch": 1.87,
2749
+ "learning_rate": 9.820839961298154e-06,
2750
+ "loss": 3.3026,
2751
+ "step": 228500
2752
+ },
2753
+ {
2754
+ "epoch": 1.88,
2755
+ "learning_rate": 9.820429984092886e-06,
2756
+ "loss": 3.301,
2757
+ "step": 229000
2758
+ },
2759
+ {
2760
+ "epoch": 1.88,
2761
+ "learning_rate": 9.820020006887618e-06,
2762
+ "loss": 3.2846,
2763
+ "step": 229500
2764
+ },
2765
+ {
2766
+ "epoch": 1.88,
2767
+ "learning_rate": 9.819610029682351e-06,
2768
+ "loss": 3.2885,
2769
+ "step": 230000
2770
+ },
2771
+ {
2772
+ "epoch": 1.89,
2773
+ "learning_rate": 9.819200052477082e-06,
2774
+ "loss": 3.2807,
2775
+ "step": 230500
2776
+ },
2777
+ {
2778
+ "epoch": 1.89,
2779
+ "learning_rate": 9.818790075271816e-06,
2780
+ "loss": 3.2787,
2781
+ "step": 231000
2782
+ },
2783
+ {
2784
+ "epoch": 1.9,
2785
+ "learning_rate": 9.818380098066548e-06,
2786
+ "loss": 3.2746,
2787
+ "step": 231500
2788
+ },
2789
+ {
2790
+ "epoch": 1.9,
2791
+ "learning_rate": 9.81797012086128e-06,
2792
+ "loss": 3.2654,
2793
+ "step": 232000
2794
+ },
2795
+ {
2796
+ "epoch": 1.9,
2797
+ "learning_rate": 9.817560143656013e-06,
2798
+ "loss": 3.2704,
2799
+ "step": 232500
2800
+ },
2801
+ {
2802
+ "epoch": 1.91,
2803
+ "learning_rate": 9.817150166450746e-06,
2804
+ "loss": 3.2666,
2805
+ "step": 233000
2806
+ },
2807
+ {
2808
+ "epoch": 1.91,
2809
+ "learning_rate": 9.816740189245478e-06,
2810
+ "loss": 3.2688,
2811
+ "step": 233500
2812
+ },
2813
+ {
2814
+ "epoch": 1.92,
2815
+ "learning_rate": 9.816330212040212e-06,
2816
+ "loss": 3.2705,
2817
+ "step": 234000
2818
+ },
2819
+ {
2820
+ "epoch": 1.92,
2821
+ "learning_rate": 9.815920234834943e-06,
2822
+ "loss": 3.2563,
2823
+ "step": 234500
2824
+ },
2825
+ {
2826
+ "epoch": 1.93,
2827
+ "learning_rate": 9.815510257629677e-06,
2828
+ "loss": 3.2619,
2829
+ "step": 235000
2830
+ },
2831
+ {
2832
+ "epoch": 1.93,
2833
+ "learning_rate": 9.81510028042441e-06,
2834
+ "loss": 3.2557,
2835
+ "step": 235500
2836
+ },
2837
+ {
2838
+ "epoch": 1.93,
2839
+ "learning_rate": 9.814690303219142e-06,
2840
+ "loss": 3.2419,
2841
+ "step": 236000
2842
+ },
2843
+ {
2844
+ "epoch": 1.94,
2845
+ "learning_rate": 9.814280326013875e-06,
2846
+ "loss": 3.2356,
2847
+ "step": 236500
2848
+ },
2849
+ {
2850
+ "epoch": 1.94,
2851
+ "learning_rate": 9.813870348808607e-06,
2852
+ "loss": 3.2401,
2853
+ "step": 237000
2854
+ },
2855
+ {
2856
+ "epoch": 1.95,
2857
+ "learning_rate": 9.81346037160334e-06,
2858
+ "loss": 3.2418,
2859
+ "step": 237500
2860
+ },
2861
+ {
2862
+ "epoch": 1.95,
2863
+ "learning_rate": 9.813050394398072e-06,
2864
+ "loss": 3.2315,
2865
+ "step": 238000
2866
+ },
2867
+ {
2868
+ "epoch": 1.95,
2869
+ "learning_rate": 9.812640417192804e-06,
2870
+ "loss": 3.233,
2871
+ "step": 238500
2872
+ },
2873
+ {
2874
+ "epoch": 1.96,
2875
+ "learning_rate": 9.812230439987539e-06,
2876
+ "loss": 3.2181,
2877
+ "step": 239000
2878
+ },
2879
+ {
2880
+ "epoch": 1.96,
2881
+ "learning_rate": 9.81182046278227e-06,
2882
+ "loss": 3.2364,
2883
+ "step": 239500
2884
+ },
2885
+ {
2886
+ "epoch": 1.97,
2887
+ "learning_rate": 9.811410485577004e-06,
2888
+ "loss": 3.2149,
2889
+ "step": 240000
2890
+ },
2891
+ {
2892
+ "epoch": 1.97,
2893
+ "learning_rate": 9.811000508371736e-06,
2894
+ "loss": 3.2026,
2895
+ "step": 240500
2896
+ },
2897
+ {
2898
+ "epoch": 1.97,
2899
+ "learning_rate": 9.810590531166468e-06,
2900
+ "loss": 3.2053,
2901
+ "step": 241000
2902
+ },
2903
+ {
2904
+ "epoch": 1.98,
2905
+ "learning_rate": 9.810180553961201e-06,
2906
+ "loss": 3.1994,
2907
+ "step": 241500
2908
+ },
2909
+ {
2910
+ "epoch": 1.98,
2911
+ "learning_rate": 9.809770576755933e-06,
2912
+ "loss": 3.1933,
2913
+ "step": 242000
2914
+ },
2915
+ {
2916
+ "epoch": 1.99,
2917
+ "learning_rate": 9.809360599550666e-06,
2918
+ "loss": 3.1945,
2919
+ "step": 242500
2920
+ },
2921
+ {
2922
+ "epoch": 1.99,
2923
+ "learning_rate": 9.808950622345398e-06,
2924
+ "loss": 3.1907,
2925
+ "step": 243000
2926
+ },
2927
+ {
2928
+ "epoch": 1.99,
2929
+ "learning_rate": 9.80854064514013e-06,
2930
+ "loss": 3.1865,
2931
+ "step": 243500
2932
+ },
2933
+ {
2934
+ "epoch": 2.0,
2935
+ "learning_rate": 9.808130667934863e-06,
2936
+ "loss": 3.1886,
2937
+ "step": 244000
2938
+ },
2939
+ {
2940
+ "epoch": 2.0,
2941
+ "learning_rate": 9.807720690729596e-06,
2942
+ "loss": 3.1766,
2943
+ "step": 244500
2944
+ },
2945
+ {
2946
+ "epoch": 2.01,
2947
+ "learning_rate": 9.807310713524328e-06,
2948
+ "loss": 3.1757,
2949
+ "step": 245000
2950
+ },
2951
+ {
2952
+ "epoch": 2.01,
2953
+ "learning_rate": 9.806900736319062e-06,
2954
+ "loss": 3.1759,
2955
+ "step": 245500
2956
+ },
2957
+ {
2958
+ "epoch": 2.02,
2959
+ "learning_rate": 9.806490759113793e-06,
2960
+ "loss": 3.1687,
2961
+ "step": 246000
2962
+ },
2963
+ {
2964
+ "epoch": 2.02,
2965
+ "learning_rate": 9.806080781908527e-06,
2966
+ "loss": 3.1521,
2967
+ "step": 246500
2968
+ },
2969
+ {
2970
+ "epoch": 2.02,
2971
+ "learning_rate": 9.80567080470326e-06,
2972
+ "loss": 3.1649,
2973
+ "step": 247000
2974
+ },
2975
+ {
2976
+ "epoch": 2.03,
2977
+ "learning_rate": 9.805260827497992e-06,
2978
+ "loss": 3.1577,
2979
+ "step": 247500
2980
+ },
2981
+ {
2982
+ "epoch": 2.03,
2983
+ "learning_rate": 9.804850850292725e-06,
2984
+ "loss": 3.1544,
2985
+ "step": 248000
2986
+ },
2987
+ {
2988
+ "epoch": 2.04,
2989
+ "learning_rate": 9.804440873087457e-06,
2990
+ "loss": 3.1526,
2991
+ "step": 248500
2992
+ },
2993
+ {
2994
+ "epoch": 2.04,
2995
+ "learning_rate": 9.80403089588219e-06,
2996
+ "loss": 3.1477,
2997
+ "step": 249000
2998
+ },
2999
+ {
3000
+ "epoch": 2.04,
3001
+ "learning_rate": 9.803620918676922e-06,
3002
+ "loss": 3.1447,
3003
+ "step": 249500
3004
+ },
3005
+ {
3006
+ "epoch": 2.05,
3007
+ "learning_rate": 9.803210941471654e-06,
3008
+ "loss": 3.1419,
3009
+ "step": 250000
3010
+ },
3011
+ {
3012
+ "epoch": 2.05,
3013
+ "learning_rate": 9.802800964266389e-06,
3014
+ "loss": 3.1377,
3015
+ "step": 250500
3016
+ },
3017
+ {
3018
+ "epoch": 2.06,
3019
+ "learning_rate": 9.80239098706112e-06,
3020
+ "loss": 3.1521,
3021
+ "step": 251000
3022
+ },
3023
+ {
3024
+ "epoch": 2.06,
3025
+ "learning_rate": 9.801981009855853e-06,
3026
+ "loss": 3.1341,
3027
+ "step": 251500
3028
+ },
3029
+ {
3030
+ "epoch": 2.06,
3031
+ "learning_rate": 9.801571032650586e-06,
3032
+ "loss": 3.1279,
3033
+ "step": 252000
3034
+ },
3035
+ {
3036
+ "epoch": 2.07,
3037
+ "learning_rate": 9.801161055445318e-06,
3038
+ "loss": 3.1296,
3039
+ "step": 252500
3040
+ },
3041
+ {
3042
+ "epoch": 2.07,
3043
+ "learning_rate": 9.800751078240051e-06,
3044
+ "loss": 3.1243,
3045
+ "step": 253000
3046
+ },
3047
+ {
3048
+ "epoch": 2.08,
3049
+ "learning_rate": 9.800341101034783e-06,
3050
+ "loss": 3.1164,
3051
+ "step": 253500
3052
+ },
3053
+ {
3054
+ "epoch": 2.08,
3055
+ "learning_rate": 9.799931123829516e-06,
3056
+ "loss": 3.1177,
3057
+ "step": 254000
3058
+ },
3059
+ {
3060
+ "epoch": 2.09,
3061
+ "learning_rate": 9.799521146624248e-06,
3062
+ "loss": 3.1196,
3063
+ "step": 254500
3064
+ },
3065
+ {
3066
+ "epoch": 2.09,
3067
+ "learning_rate": 9.79911116941898e-06,
3068
+ "loss": 3.1148,
3069
+ "step": 255000
3070
+ },
3071
+ {
3072
+ "epoch": 2.09,
3073
+ "learning_rate": 9.798701192213713e-06,
3074
+ "loss": 3.1061,
3075
+ "step": 255500
3076
+ },
3077
+ {
3078
+ "epoch": 2.1,
3079
+ "learning_rate": 9.798291215008446e-06,
3080
+ "loss": 3.1101,
3081
+ "step": 256000
3082
+ },
3083
+ {
3084
+ "epoch": 2.1,
3085
+ "learning_rate": 9.797881237803178e-06,
3086
+ "loss": 3.107,
3087
+ "step": 256500
3088
+ },
3089
+ {
3090
+ "epoch": 2.11,
3091
+ "learning_rate": 9.797471260597912e-06,
3092
+ "loss": 3.1068,
3093
+ "step": 257000
3094
+ },
3095
+ {
3096
+ "epoch": 2.11,
3097
+ "learning_rate": 9.797061283392643e-06,
3098
+ "loss": 3.108,
3099
+ "step": 257500
3100
+ },
3101
+ {
3102
+ "epoch": 2.11,
3103
+ "learning_rate": 9.796651306187377e-06,
3104
+ "loss": 3.0922,
3105
+ "step": 258000
3106
+ },
3107
+ {
3108
+ "epoch": 2.12,
3109
+ "learning_rate": 9.79624132898211e-06,
3110
+ "loss": 3.0889,
3111
+ "step": 258500
3112
+ },
3113
+ {
3114
+ "epoch": 2.12,
3115
+ "learning_rate": 9.795831351776842e-06,
3116
+ "loss": 3.0981,
3117
+ "step": 259000
3118
+ },
3119
+ {
3120
+ "epoch": 2.13,
3121
+ "learning_rate": 9.795421374571575e-06,
3122
+ "loss": 3.0861,
3123
+ "step": 259500
3124
+ },
3125
+ {
3126
+ "epoch": 2.13,
3127
+ "learning_rate": 9.795011397366307e-06,
3128
+ "loss": 3.0924,
3129
+ "step": 260000
3130
+ },
3131
+ {
3132
+ "epoch": 2.13,
3133
+ "learning_rate": 9.79460142016104e-06,
3134
+ "loss": 3.0889,
3135
+ "step": 260500
3136
+ },
3137
+ {
3138
+ "epoch": 2.14,
3139
+ "learning_rate": 9.794191442955772e-06,
3140
+ "loss": 3.0936,
3141
+ "step": 261000
3142
+ },
3143
+ {
3144
+ "epoch": 2.14,
3145
+ "learning_rate": 9.793781465750504e-06,
3146
+ "loss": 3.0831,
3147
+ "step": 261500
3148
+ },
3149
+ {
3150
+ "epoch": 2.15,
3151
+ "learning_rate": 9.793371488545239e-06,
3152
+ "loss": 3.0684,
3153
+ "step": 262000
3154
+ },
3155
+ {
3156
+ "epoch": 2.15,
3157
+ "learning_rate": 9.79296151133997e-06,
3158
+ "loss": 3.078,
3159
+ "step": 262500
3160
+ },
3161
+ {
3162
+ "epoch": 2.15,
3163
+ "learning_rate": 9.792551534134703e-06,
3164
+ "loss": 3.0753,
3165
+ "step": 263000
3166
+ },
3167
+ {
3168
+ "epoch": 2.16,
3169
+ "learning_rate": 9.792141556929436e-06,
3170
+ "loss": 3.0687,
3171
+ "step": 263500
3172
+ },
3173
+ {
3174
+ "epoch": 2.16,
3175
+ "learning_rate": 9.791731579724168e-06,
3176
+ "loss": 3.0662,
3177
+ "step": 264000
3178
+ },
3179
+ {
3180
+ "epoch": 2.17,
3181
+ "learning_rate": 9.7913216025189e-06,
3182
+ "loss": 3.0591,
3183
+ "step": 264500
3184
+ },
3185
+ {
3186
+ "epoch": 2.17,
3187
+ "learning_rate": 9.790911625313633e-06,
3188
+ "loss": 3.06,
3189
+ "step": 265000
3190
+ },
3191
+ {
3192
+ "epoch": 2.18,
3193
+ "learning_rate": 9.790501648108366e-06,
3194
+ "loss": 3.0507,
3195
+ "step": 265500
3196
+ },
3197
+ {
3198
+ "epoch": 2.18,
3199
+ "learning_rate": 9.7900916709031e-06,
3200
+ "loss": 3.0593,
3201
+ "step": 266000
3202
+ },
3203
+ {
3204
+ "epoch": 2.18,
3205
+ "learning_rate": 9.78968169369783e-06,
3206
+ "loss": 3.0589,
3207
+ "step": 266500
3208
+ },
3209
+ {
3210
+ "epoch": 2.19,
3211
+ "learning_rate": 9.789271716492565e-06,
3212
+ "loss": 3.0523,
3213
+ "step": 267000
3214
+ },
3215
+ {
3216
+ "epoch": 2.19,
3217
+ "learning_rate": 9.788861739287297e-06,
3218
+ "loss": 3.0443,
3219
+ "step": 267500
3220
+ },
3221
+ {
3222
+ "epoch": 2.2,
3223
+ "learning_rate": 9.788451762082028e-06,
3224
+ "loss": 3.0464,
3225
+ "step": 268000
3226
+ },
3227
+ {
3228
+ "epoch": 2.2,
3229
+ "learning_rate": 9.788041784876762e-06,
3230
+ "loss": 3.0398,
3231
+ "step": 268500
3232
+ },
3233
+ {
3234
+ "epoch": 2.2,
3235
+ "learning_rate": 9.787631807671493e-06,
3236
+ "loss": 3.0371,
3237
+ "step": 269000
3238
+ },
3239
+ {
3240
+ "epoch": 2.21,
3241
+ "learning_rate": 9.787221830466227e-06,
3242
+ "loss": 3.0287,
3243
+ "step": 269500
3244
+ },
3245
+ {
3246
+ "epoch": 2.21,
3247
+ "learning_rate": 9.78681185326096e-06,
3248
+ "loss": 3.0457,
3249
+ "step": 270000
3250
+ },
3251
+ {
3252
+ "epoch": 2.22,
3253
+ "learning_rate": 9.786401876055692e-06,
3254
+ "loss": 3.0355,
3255
+ "step": 270500
3256
+ },
3257
+ {
3258
+ "epoch": 2.22,
3259
+ "learning_rate": 9.785991898850424e-06,
3260
+ "loss": 3.0385,
3261
+ "step": 271000
3262
+ },
3263
+ {
3264
+ "epoch": 2.22,
3265
+ "learning_rate": 9.785581921645157e-06,
3266
+ "loss": 3.0242,
3267
+ "step": 271500
3268
+ },
3269
+ {
3270
+ "epoch": 2.23,
3271
+ "learning_rate": 9.78517194443989e-06,
3272
+ "loss": 3.0248,
3273
+ "step": 272000
3274
+ },
3275
+ {
3276
+ "epoch": 2.23,
3277
+ "learning_rate": 9.784761967234624e-06,
3278
+ "loss": 3.0148,
3279
+ "step": 272500
3280
+ },
3281
+ {
3282
+ "epoch": 2.24,
3283
+ "learning_rate": 9.784351990029354e-06,
3284
+ "loss": 3.0449,
3285
+ "step": 273000
3286
+ },
3287
+ {
3288
+ "epoch": 2.24,
3289
+ "learning_rate": 9.783942012824088e-06,
3290
+ "loss": 3.0311,
3291
+ "step": 273500
3292
+ },
3293
+ {
3294
+ "epoch": 2.24,
3295
+ "learning_rate": 9.783532035618821e-06,
3296
+ "loss": 3.0237,
3297
+ "step": 274000
3298
+ },
3299
+ {
3300
+ "epoch": 2.25,
3301
+ "learning_rate": 9.783122058413553e-06,
3302
+ "loss": 3.0253,
3303
+ "step": 274500
3304
+ },
3305
+ {
3306
+ "epoch": 2.25,
3307
+ "learning_rate": 9.782712081208286e-06,
3308
+ "loss": 3.0234,
3309
+ "step": 275000
3310
+ },
3311
+ {
3312
+ "epoch": 2.26,
3313
+ "learning_rate": 9.782302104003018e-06,
3314
+ "loss": 3.0212,
3315
+ "step": 275500
3316
+ },
3317
+ {
3318
+ "epoch": 2.26,
3319
+ "learning_rate": 9.78189212679775e-06,
3320
+ "loss": 3.0142,
3321
+ "step": 276000
3322
+ },
3323
+ {
3324
+ "epoch": 2.27,
3325
+ "learning_rate": 9.781482149592483e-06,
3326
+ "loss": 3.0159,
3327
+ "step": 276500
3328
+ },
3329
+ {
3330
+ "epoch": 2.27,
3331
+ "learning_rate": 9.781072172387216e-06,
3332
+ "loss": 3.0111,
3333
+ "step": 277000
3334
+ },
3335
+ {
3336
+ "epoch": 2.27,
3337
+ "learning_rate": 9.78066219518195e-06,
3338
+ "loss": 3.0126,
3339
+ "step": 277500
3340
+ },
3341
+ {
3342
+ "epoch": 2.28,
3343
+ "learning_rate": 9.78025221797668e-06,
3344
+ "loss": 3.0104,
3345
+ "step": 278000
3346
+ },
3347
+ {
3348
+ "epoch": 2.28,
3349
+ "learning_rate": 9.779842240771415e-06,
3350
+ "loss": 2.9987,
3351
+ "step": 278500
3352
+ },
3353
+ {
3354
+ "epoch": 2.29,
3355
+ "learning_rate": 9.779432263566147e-06,
3356
+ "loss": 2.994,
3357
+ "step": 279000
3358
+ },
3359
+ {
3360
+ "epoch": 2.29,
3361
+ "learning_rate": 9.779022286360878e-06,
3362
+ "loss": 2.9956,
3363
+ "step": 279500
3364
+ },
3365
+ {
3366
+ "epoch": 2.29,
3367
+ "learning_rate": 9.778612309155612e-06,
3368
+ "loss": 2.9843,
3369
+ "step": 280000
3370
+ },
3371
+ {
3372
+ "epoch": 2.3,
3373
+ "learning_rate": 9.778202331950345e-06,
3374
+ "loss": 2.9861,
3375
+ "step": 280500
3376
+ },
3377
+ {
3378
+ "epoch": 2.3,
3379
+ "learning_rate": 9.777792354745077e-06,
3380
+ "loss": 2.992,
3381
+ "step": 281000
3382
+ },
3383
+ {
3384
+ "epoch": 2.31,
3385
+ "learning_rate": 9.77738237753981e-06,
3386
+ "loss": 2.989,
3387
+ "step": 281500
3388
+ },
3389
+ {
3390
+ "epoch": 2.31,
3391
+ "learning_rate": 9.776972400334542e-06,
3392
+ "loss": 2.9962,
3393
+ "step": 282000
3394
+ },
3395
+ {
3396
+ "epoch": 2.31,
3397
+ "learning_rate": 9.776562423129274e-06,
3398
+ "loss": 2.9844,
3399
+ "step": 282500
3400
+ },
3401
+ {
3402
+ "epoch": 2.32,
3403
+ "learning_rate": 9.776152445924007e-06,
3404
+ "loss": 2.9777,
3405
+ "step": 283000
3406
+ },
3407
+ {
3408
+ "epoch": 2.32,
3409
+ "learning_rate": 9.77574246871874e-06,
3410
+ "loss": 2.9806,
3411
+ "step": 283500
3412
+ },
3413
+ {
3414
+ "epoch": 2.33,
3415
+ "learning_rate": 9.775332491513474e-06,
3416
+ "loss": 2.9783,
3417
+ "step": 284000
3418
+ },
3419
+ {
3420
+ "epoch": 2.33,
3421
+ "learning_rate": 9.774922514308204e-06,
3422
+ "loss": 2.9717,
3423
+ "step": 284500
3424
+ },
3425
+ {
3426
+ "epoch": 2.33,
3427
+ "learning_rate": 9.774512537102938e-06,
3428
+ "loss": 2.9753,
3429
+ "step": 285000
3430
+ },
3431
+ {
3432
+ "epoch": 2.34,
3433
+ "learning_rate": 9.774102559897671e-06,
3434
+ "loss": 3.0081,
3435
+ "step": 285500
3436
+ },
3437
+ {
3438
+ "epoch": 2.34,
3439
+ "learning_rate": 9.773692582692403e-06,
3440
+ "loss": 2.9719,
3441
+ "step": 286000
3442
+ },
3443
+ {
3444
+ "epoch": 2.35,
3445
+ "learning_rate": 9.773282605487136e-06,
3446
+ "loss": 2.9699,
3447
+ "step": 286500
3448
+ },
3449
+ {
3450
+ "epoch": 2.35,
3451
+ "learning_rate": 9.772872628281868e-06,
3452
+ "loss": 2.9652,
3453
+ "step": 287000
3454
+ },
3455
+ {
3456
+ "epoch": 2.36,
3457
+ "learning_rate": 9.7724626510766e-06,
3458
+ "loss": 2.9697,
3459
+ "step": 287500
3460
+ },
3461
+ {
3462
+ "epoch": 2.36,
3463
+ "learning_rate": 9.772052673871333e-06,
3464
+ "loss": 2.9538,
3465
+ "step": 288000
3466
+ },
3467
+ {
3468
+ "epoch": 2.36,
3469
+ "learning_rate": 9.771642696666066e-06,
3470
+ "loss": 2.9509,
3471
+ "step": 288500
3472
+ },
3473
+ {
3474
+ "epoch": 2.37,
3475
+ "learning_rate": 9.7712327194608e-06,
3476
+ "loss": 2.954,
3477
+ "step": 289000
3478
+ },
3479
+ {
3480
+ "epoch": 2.37,
3481
+ "learning_rate": 9.77082274225553e-06,
3482
+ "loss": 2.9578,
3483
+ "step": 289500
3484
+ },
3485
+ {
3486
+ "epoch": 2.38,
3487
+ "learning_rate": 9.770412765050265e-06,
3488
+ "loss": 2.9521,
3489
+ "step": 290000
3490
+ },
3491
+ {
3492
+ "epoch": 2.38,
3493
+ "learning_rate": 9.770002787844997e-06,
3494
+ "loss": 2.9494,
3495
+ "step": 290500
3496
+ },
3497
+ {
3498
+ "epoch": 2.38,
3499
+ "learning_rate": 9.76959281063973e-06,
3500
+ "loss": 2.9559,
3501
+ "step": 291000
3502
+ },
3503
+ {
3504
+ "epoch": 2.39,
3505
+ "learning_rate": 9.769182833434462e-06,
3506
+ "loss": 2.9464,
3507
+ "step": 291500
3508
+ },
3509
+ {
3510
+ "epoch": 2.39,
3511
+ "learning_rate": 9.768772856229195e-06,
3512
+ "loss": 2.9491,
3513
+ "step": 292000
3514
+ },
3515
+ {
3516
+ "epoch": 2.4,
3517
+ "learning_rate": 9.768362879023927e-06,
3518
+ "loss": 2.946,
3519
+ "step": 292500
3520
+ },
3521
+ {
3522
+ "epoch": 2.4,
3523
+ "step": 292842,
3524
+ "total_flos": 2.3285464173306184e+18,
3525
+ "train_loss": 4.907397945397096,
3526
+ "train_runtime": 197998.9762,
3527
+ "train_samples_per_second": 2958.984,
3528
+ "train_steps_per_second": 61.646
3529
+ }
3530
+ ],
3531
+ "logging_steps": 500,
3532
+ "max_steps": 12205800,
3533
+ "num_train_epochs": 100,
3534
+ "save_steps": 1000000,
3535
+ "total_flos": 2.3285464173306184e+18,
3536
+ "trial_name": null,
3537
+ "trial_params": null
3538
+ }