Zhou Yucheng commited on
Commit
b247ff3
β€’
1 Parent(s): e3d924b

First model version

Browse files
README.md CHANGED
@@ -1,3 +1,83 @@
1
  ---
2
  license: apache-2.0
 
 
 
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: apache-2.0
3
+ language: en
4
+ tags:
5
+ - generated_from_trainer
6
+ metrics:
7
+ - precision
8
+ - recall
9
+ - f1
10
+ - accuracy
11
+ model-index:
12
+ - name: macbert-finetuned-tokenclassification-errorword
13
+ results: []
14
  ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # macbert-finetuned-tokenclassification-errorword
20
+
21
+ This model is a fine-tuned version of [shibing624/macbert4csc-base-chinese](https://huggingface.co/shibing624/macbert4csc-base-chinese) on an unknown dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 0.0040
24
+ - Precision: 0.0
25
+ - Recall: 0.0
26
+ - F1: 0.0
27
+ - Accuracy: 0.9994
28
+
29
+ ## Model description
30
+
31
+ This model fine-tuned on a large corpus of medical material which processed on purpose, we propose to sample words and use similar words to do replacement for masking purpose.
32
+ As a result, this model can performed pretty well when applying on medical relatted downstream tasks.
33
+
34
+ ## Intended uses & limitations
35
+
36
+ You can use this model directly with a pipeline for token classification:
37
+ ```python
38
+ from transformers import (AutoModelForTokenClassification, AutoTokenizer
39
+ from transformers import pipeline
40
+
41
+ hub_model_id = "9pinus/macbert-base-chinese-medical-collation"
42
+
43
+ model = AutoModelForTokenClassification.from_pretrained(hub_model_id)
44
+ tokenizer = BertTokenizer.from_pretrained(hub_model_id)
45
+ classifier = pipeline('ner', model=model, tokenizer=tokenizer)
46
+ result = classifier("ε¦‚ζžœη—…ζƒ…θΎƒι‡οΌŒε―ι€‚ε½“ε£ζœη”²η‘ε”‘η‰‡γ€ηŽ―ι…―ηΊ’ιœ‰η΄ η‰‡γ€ε²ε“šηΎŽθΎ›η‰‡η­‰θ―η‰©θΏ›θ‘ŒζŠ—ζ„ŸζŸ“ι•‡η—›γ€‚εŒζ—Άεœ¨ζ—₯εΈΈη”Ÿζ΄»δΈ­θ¦ζ³¨ζ„η‰™ι½ΏζΈ…ζ΄ε«η”ŸοΌŒε…»ζˆεˆ·η‰™ηš„ε₯½δΉ ζƒ―。")
47
+
48
+ for item in result:
49
+ print(item)
50
+ ```
51
+
52
+ ### Training hyperparameters
53
+
54
+ The following hyperparameters were used during training:
55
+ - learning_rate: 5e-05
56
+ - train_batch_size: 16
57
+ - eval_batch_size: 16
58
+ - seed: 42
59
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
+ - lr_scheduler_type: linear
61
+ - num_epochs: 8.0
62
+ - mixed_precision_training: Native AMP
63
+
64
+ ### Training results
65
+
66
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
67
+ |:-------------:|:-----:|:------:|:---------------:|:---------:|:------:|:---:|:--------:|
68
+ | 0.0038 | 1.0 | 36875 | 0.0030 | 0.0 | 0.0 | 0.0 | 0.9991 |
69
+ | 0.0026 | 2.0 | 73750 | 0.0028 | 0.0 | 0.0 | 0.0 | 0.9992 |
70
+ | 0.0021 | 3.0 | 110625 | 0.0033 | 0.0 | 0.0 | 0.0 | 0.9992 |
71
+ | 0.0014 | 4.0 | 147500 | 0.0033 | 0.0 | 0.0 | 0.0 | 0.9993 |
72
+ | 0.0009 | 5.0 | 184375 | 0.0033 | 0.0 | 0.0 | 0.0 | 0.9993 |
73
+ | 0.0006 | 6.0 | 221250 | 0.0035 | 0.0 | 0.0 | 0.0 | 0.9994 |
74
+ | 0.0004 | 7.0 | 258125 | 0.0037 | 0.0 | 0.0 | 0.0 | 0.9994 |
75
+ | 0.0002 | 8.0 | 295000 | 0.0040 | 0.0 | 0.0 | 0.0 | 0.9994 |
76
+
77
+
78
+ ### Framework versions
79
+
80
+ - Transformers 4.15.0
81
+ - Pytorch 1.10.1+cu113
82
+ - Datasets 1.17.0
83
+ - Tokenizers 0.10.3
eval_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 8.0,
3
+ "eval_accuracy": 0.9994014164827584,
4
+ "eval_f1": 0.0,
5
+ "eval_loss": 0.003991865552961826,
6
+ "eval_precision": 0.0,
7
+ "eval_recall": 0.0,
8
+ "eval_runtime": 375.9642,
9
+ "eval_samples": 39609,
10
+ "eval_samples_per_second": 105.353,
11
+ "eval_steps_per_second": 6.586
12
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f693cf29610097f978392e17ae017670dc9ebc0921c32922df56b524b6608578
3
+ size 406926833
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "do_basic_tokenize": true, "never_split": null, "name_or_path": "shibing624/macbert4csc-base-chinese", "special_tokens_map_file": "cache\\5559a99a8a2784cef7bf4784ffdb8f0ec2df952f0ca49d9aa3c7f46530fbc8ef.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d", "tokenizer_class": "BertTokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 8.0,
3
+ "train_loss": 0.0015094334459279553,
4
+ "train_runtime": 87513.7882,
5
+ "train_samples": 590000,
6
+ "train_samples_per_second": 53.934,
7
+ "train_steps_per_second": 3.371
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,3661 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 8.0,
5
+ "global_step": 295000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 4.991559322033899e-05,
13
+ "loss": 0.0122,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.03,
18
+ "learning_rate": 4.983084745762712e-05,
19
+ "loss": 0.0065,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.04,
24
+ "learning_rate": 4.974610169491526e-05,
25
+ "loss": 0.0059,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.05,
30
+ "learning_rate": 4.966135593220339e-05,
31
+ "loss": 0.0056,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.07,
36
+ "learning_rate": 4.9576610169491526e-05,
37
+ "loss": 0.0053,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.08,
42
+ "learning_rate": 4.9491864406779666e-05,
43
+ "loss": 0.0058,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.09,
48
+ "learning_rate": 4.94071186440678e-05,
49
+ "loss": 0.0054,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.11,
54
+ "learning_rate": 4.932237288135593e-05,
55
+ "loss": 0.0049,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.12,
60
+ "learning_rate": 4.923762711864407e-05,
61
+ "loss": 0.0053,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.14,
66
+ "learning_rate": 4.9152881355932206e-05,
67
+ "loss": 0.005,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.15,
72
+ "learning_rate": 4.9068135593220346e-05,
73
+ "loss": 0.0051,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.16,
78
+ "learning_rate": 4.898338983050847e-05,
79
+ "loss": 0.0049,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.18,
84
+ "learning_rate": 4.889864406779661e-05,
85
+ "loss": 0.0046,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.19,
90
+ "learning_rate": 4.881389830508475e-05,
91
+ "loss": 0.005,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.2,
96
+ "learning_rate": 4.8729152542372885e-05,
97
+ "loss": 0.0046,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.22,
102
+ "learning_rate": 4.864440677966102e-05,
103
+ "loss": 0.0046,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.23,
108
+ "learning_rate": 4.855966101694915e-05,
109
+ "loss": 0.0045,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.24,
114
+ "learning_rate": 4.847491525423729e-05,
115
+ "loss": 0.0046,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.26,
120
+ "learning_rate": 4.8390169491525425e-05,
121
+ "loss": 0.0046,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 0.27,
126
+ "learning_rate": 4.830542372881356e-05,
127
+ "loss": 0.0044,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 0.28,
132
+ "learning_rate": 4.82206779661017e-05,
133
+ "loss": 0.0046,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 0.3,
138
+ "learning_rate": 4.813593220338983e-05,
139
+ "loss": 0.0043,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 0.31,
144
+ "learning_rate": 4.805118644067797e-05,
145
+ "loss": 0.0042,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 0.33,
150
+ "learning_rate": 4.7966440677966104e-05,
151
+ "loss": 0.0045,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 0.34,
156
+ "learning_rate": 4.7881864406779666e-05,
157
+ "loss": 0.0044,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 0.35,
162
+ "learning_rate": 4.77971186440678e-05,
163
+ "loss": 0.0046,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 0.37,
168
+ "learning_rate": 4.771237288135593e-05,
169
+ "loss": 0.0043,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 0.38,
174
+ "learning_rate": 4.7627627118644066e-05,
175
+ "loss": 0.0043,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 0.39,
180
+ "learning_rate": 4.7542881355932206e-05,
181
+ "loss": 0.0045,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 0.41,
186
+ "learning_rate": 4.745830508474576e-05,
187
+ "loss": 0.0041,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 0.42,
192
+ "learning_rate": 4.73735593220339e-05,
193
+ "loss": 0.0041,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 0.43,
198
+ "learning_rate": 4.7288813559322034e-05,
199
+ "loss": 0.0044,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 0.45,
204
+ "learning_rate": 4.7204067796610174e-05,
205
+ "loss": 0.0044,
206
+ "step": 16500
207
+ },
208
+ {
209
+ "epoch": 0.46,
210
+ "learning_rate": 4.711949152542373e-05,
211
+ "loss": 0.0044,
212
+ "step": 17000
213
+ },
214
+ {
215
+ "epoch": 0.47,
216
+ "learning_rate": 4.703474576271187e-05,
217
+ "loss": 0.004,
218
+ "step": 17500
219
+ },
220
+ {
221
+ "epoch": 0.49,
222
+ "learning_rate": 4.695e-05,
223
+ "loss": 0.0041,
224
+ "step": 18000
225
+ },
226
+ {
227
+ "epoch": 0.5,
228
+ "learning_rate": 4.6865423728813565e-05,
229
+ "loss": 0.004,
230
+ "step": 18500
231
+ },
232
+ {
233
+ "epoch": 0.52,
234
+ "learning_rate": 4.67806779661017e-05,
235
+ "loss": 0.0039,
236
+ "step": 19000
237
+ },
238
+ {
239
+ "epoch": 0.53,
240
+ "learning_rate": 4.669593220338983e-05,
241
+ "loss": 0.004,
242
+ "step": 19500
243
+ },
244
+ {
245
+ "epoch": 0.54,
246
+ "learning_rate": 4.6611186440677964e-05,
247
+ "loss": 0.0041,
248
+ "step": 20000
249
+ },
250
+ {
251
+ "epoch": 0.56,
252
+ "learning_rate": 4.6526440677966104e-05,
253
+ "loss": 0.004,
254
+ "step": 20500
255
+ },
256
+ {
257
+ "epoch": 0.57,
258
+ "learning_rate": 4.644186440677966e-05,
259
+ "loss": 0.0038,
260
+ "step": 21000
261
+ },
262
+ {
263
+ "epoch": 0.58,
264
+ "learning_rate": 4.635728813559322e-05,
265
+ "loss": 0.0041,
266
+ "step": 21500
267
+ },
268
+ {
269
+ "epoch": 0.6,
270
+ "learning_rate": 4.6272542372881355e-05,
271
+ "loss": 0.0037,
272
+ "step": 22000
273
+ },
274
+ {
275
+ "epoch": 0.61,
276
+ "learning_rate": 4.6187796610169495e-05,
277
+ "loss": 0.0039,
278
+ "step": 22500
279
+ },
280
+ {
281
+ "epoch": 0.62,
282
+ "learning_rate": 4.610305084745763e-05,
283
+ "loss": 0.004,
284
+ "step": 23000
285
+ },
286
+ {
287
+ "epoch": 0.64,
288
+ "learning_rate": 4.601830508474577e-05,
289
+ "loss": 0.0041,
290
+ "step": 23500
291
+ },
292
+ {
293
+ "epoch": 0.65,
294
+ "learning_rate": 4.5933559322033895e-05,
295
+ "loss": 0.0037,
296
+ "step": 24000
297
+ },
298
+ {
299
+ "epoch": 0.66,
300
+ "learning_rate": 4.5848983050847464e-05,
301
+ "loss": 0.0042,
302
+ "step": 24500
303
+ },
304
+ {
305
+ "epoch": 0.68,
306
+ "learning_rate": 4.57642372881356e-05,
307
+ "loss": 0.0041,
308
+ "step": 25000
309
+ },
310
+ {
311
+ "epoch": 0.69,
312
+ "learning_rate": 4.567949152542373e-05,
313
+ "loss": 0.004,
314
+ "step": 25500
315
+ },
316
+ {
317
+ "epoch": 0.71,
318
+ "learning_rate": 4.559474576271186e-05,
319
+ "loss": 0.0039,
320
+ "step": 26000
321
+ },
322
+ {
323
+ "epoch": 0.72,
324
+ "learning_rate": 4.551e-05,
325
+ "loss": 0.0039,
326
+ "step": 26500
327
+ },
328
+ {
329
+ "epoch": 0.73,
330
+ "learning_rate": 4.5425254237288136e-05,
331
+ "loss": 0.0038,
332
+ "step": 27000
333
+ },
334
+ {
335
+ "epoch": 0.75,
336
+ "learning_rate": 4.534050847457627e-05,
337
+ "loss": 0.0041,
338
+ "step": 27500
339
+ },
340
+ {
341
+ "epoch": 0.76,
342
+ "learning_rate": 4.525576271186441e-05,
343
+ "loss": 0.0037,
344
+ "step": 28000
345
+ },
346
+ {
347
+ "epoch": 0.77,
348
+ "learning_rate": 4.517118644067797e-05,
349
+ "loss": 0.004,
350
+ "step": 28500
351
+ },
352
+ {
353
+ "epoch": 0.79,
354
+ "learning_rate": 4.5086440677966105e-05,
355
+ "loss": 0.004,
356
+ "step": 29000
357
+ },
358
+ {
359
+ "epoch": 0.8,
360
+ "learning_rate": 4.500169491525424e-05,
361
+ "loss": 0.0042,
362
+ "step": 29500
363
+ },
364
+ {
365
+ "epoch": 0.81,
366
+ "learning_rate": 4.491694915254238e-05,
367
+ "loss": 0.004,
368
+ "step": 30000
369
+ },
370
+ {
371
+ "epoch": 0.83,
372
+ "learning_rate": 4.483220338983051e-05,
373
+ "loss": 0.0034,
374
+ "step": 30500
375
+ },
376
+ {
377
+ "epoch": 0.84,
378
+ "learning_rate": 4.474762711864407e-05,
379
+ "loss": 0.0039,
380
+ "step": 31000
381
+ },
382
+ {
383
+ "epoch": 0.85,
384
+ "learning_rate": 4.4662881355932206e-05,
385
+ "loss": 0.0037,
386
+ "step": 31500
387
+ },
388
+ {
389
+ "epoch": 0.87,
390
+ "learning_rate": 4.4578135593220346e-05,
391
+ "loss": 0.0039,
392
+ "step": 32000
393
+ },
394
+ {
395
+ "epoch": 0.88,
396
+ "learning_rate": 4.449338983050847e-05,
397
+ "loss": 0.0035,
398
+ "step": 32500
399
+ },
400
+ {
401
+ "epoch": 0.89,
402
+ "learning_rate": 4.440864406779661e-05,
403
+ "loss": 0.0041,
404
+ "step": 33000
405
+ },
406
+ {
407
+ "epoch": 0.91,
408
+ "learning_rate": 4.4323898305084746e-05,
409
+ "loss": 0.0041,
410
+ "step": 33500
411
+ },
412
+ {
413
+ "epoch": 0.92,
414
+ "learning_rate": 4.4239152542372886e-05,
415
+ "loss": 0.004,
416
+ "step": 34000
417
+ },
418
+ {
419
+ "epoch": 0.94,
420
+ "learning_rate": 4.415440677966102e-05,
421
+ "loss": 0.0038,
422
+ "step": 34500
423
+ },
424
+ {
425
+ "epoch": 0.95,
426
+ "learning_rate": 4.406983050847458e-05,
427
+ "loss": 0.0038,
428
+ "step": 35000
429
+ },
430
+ {
431
+ "epoch": 0.96,
432
+ "learning_rate": 4.3985084745762714e-05,
433
+ "loss": 0.0037,
434
+ "step": 35500
435
+ },
436
+ {
437
+ "epoch": 0.98,
438
+ "learning_rate": 4.390033898305085e-05,
439
+ "loss": 0.004,
440
+ "step": 36000
441
+ },
442
+ {
443
+ "epoch": 0.99,
444
+ "learning_rate": 4.381559322033899e-05,
445
+ "loss": 0.0038,
446
+ "step": 36500
447
+ },
448
+ {
449
+ "epoch": 1.0,
450
+ "eval_accuracy": 0.9991040475613329,
451
+ "eval_f1": 0.0,
452
+ "eval_loss": 0.0030166972428560257,
453
+ "eval_precision": 0.0,
454
+ "eval_recall": 0.0,
455
+ "eval_runtime": 368.0366,
456
+ "eval_samples_per_second": 107.622,
457
+ "eval_steps_per_second": 6.728,
458
+ "step": 36875
459
+ },
460
+ {
461
+ "epoch": 1.0,
462
+ "learning_rate": 4.373084745762712e-05,
463
+ "loss": 0.0037,
464
+ "step": 37000
465
+ },
466
+ {
467
+ "epoch": 1.02,
468
+ "learning_rate": 4.364610169491526e-05,
469
+ "loss": 0.0026,
470
+ "step": 37500
471
+ },
472
+ {
473
+ "epoch": 1.03,
474
+ "learning_rate": 4.356135593220339e-05,
475
+ "loss": 0.0026,
476
+ "step": 38000
477
+ },
478
+ {
479
+ "epoch": 1.04,
480
+ "learning_rate": 4.347661016949153e-05,
481
+ "loss": 0.0025,
482
+ "step": 38500
483
+ },
484
+ {
485
+ "epoch": 1.06,
486
+ "learning_rate": 4.339186440677967e-05,
487
+ "loss": 0.0024,
488
+ "step": 39000
489
+ },
490
+ {
491
+ "epoch": 1.07,
492
+ "learning_rate": 4.33071186440678e-05,
493
+ "loss": 0.002,
494
+ "step": 39500
495
+ },
496
+ {
497
+ "epoch": 1.08,
498
+ "learning_rate": 4.322237288135593e-05,
499
+ "loss": 0.0024,
500
+ "step": 40000
501
+ },
502
+ {
503
+ "epoch": 1.1,
504
+ "learning_rate": 4.3137796610169495e-05,
505
+ "loss": 0.0031,
506
+ "step": 40500
507
+ },
508
+ {
509
+ "epoch": 1.11,
510
+ "learning_rate": 4.305338983050848e-05,
511
+ "loss": 0.0034,
512
+ "step": 41000
513
+ },
514
+ {
515
+ "epoch": 1.13,
516
+ "learning_rate": 4.296864406779661e-05,
517
+ "loss": 0.0031,
518
+ "step": 41500
519
+ },
520
+ {
521
+ "epoch": 1.14,
522
+ "learning_rate": 4.2883898305084746e-05,
523
+ "loss": 0.0029,
524
+ "step": 42000
525
+ },
526
+ {
527
+ "epoch": 1.15,
528
+ "learning_rate": 4.279932203389831e-05,
529
+ "loss": 0.0025,
530
+ "step": 42500
531
+ },
532
+ {
533
+ "epoch": 1.17,
534
+ "learning_rate": 4.271457627118644e-05,
535
+ "loss": 0.0025,
536
+ "step": 43000
537
+ },
538
+ {
539
+ "epoch": 1.18,
540
+ "learning_rate": 4.2629830508474575e-05,
541
+ "loss": 0.0026,
542
+ "step": 43500
543
+ },
544
+ {
545
+ "epoch": 1.19,
546
+ "learning_rate": 4.2545084745762715e-05,
547
+ "loss": 0.0028,
548
+ "step": 44000
549
+ },
550
+ {
551
+ "epoch": 1.21,
552
+ "learning_rate": 4.2460338983050855e-05,
553
+ "loss": 0.0026,
554
+ "step": 44500
555
+ },
556
+ {
557
+ "epoch": 1.22,
558
+ "learning_rate": 4.237559322033898e-05,
559
+ "loss": 0.0028,
560
+ "step": 45000
561
+ },
562
+ {
563
+ "epoch": 1.23,
564
+ "learning_rate": 4.229084745762712e-05,
565
+ "loss": 0.0023,
566
+ "step": 45500
567
+ },
568
+ {
569
+ "epoch": 1.25,
570
+ "learning_rate": 4.2206101694915254e-05,
571
+ "loss": 0.0025,
572
+ "step": 46000
573
+ },
574
+ {
575
+ "epoch": 1.26,
576
+ "learning_rate": 4.2121355932203394e-05,
577
+ "loss": 0.0027,
578
+ "step": 46500
579
+ },
580
+ {
581
+ "epoch": 1.27,
582
+ "learning_rate": 4.203661016949153e-05,
583
+ "loss": 0.0024,
584
+ "step": 47000
585
+ },
586
+ {
587
+ "epoch": 1.29,
588
+ "learning_rate": 4.195186440677966e-05,
589
+ "loss": 0.0025,
590
+ "step": 47500
591
+ },
592
+ {
593
+ "epoch": 1.3,
594
+ "learning_rate": 4.18671186440678e-05,
595
+ "loss": 0.0026,
596
+ "step": 48000
597
+ },
598
+ {
599
+ "epoch": 1.32,
600
+ "learning_rate": 4.1782372881355934e-05,
601
+ "loss": 0.0028,
602
+ "step": 48500
603
+ },
604
+ {
605
+ "epoch": 1.33,
606
+ "learning_rate": 4.1697627118644074e-05,
607
+ "loss": 0.0029,
608
+ "step": 49000
609
+ },
610
+ {
611
+ "epoch": 1.34,
612
+ "learning_rate": 4.161288135593221e-05,
613
+ "loss": 0.0025,
614
+ "step": 49500
615
+ },
616
+ {
617
+ "epoch": 1.36,
618
+ "learning_rate": 4.152813559322034e-05,
619
+ "loss": 0.0024,
620
+ "step": 50000
621
+ },
622
+ {
623
+ "epoch": 1.37,
624
+ "learning_rate": 4.144338983050848e-05,
625
+ "loss": 0.0025,
626
+ "step": 50500
627
+ },
628
+ {
629
+ "epoch": 1.38,
630
+ "learning_rate": 4.135864406779661e-05,
631
+ "loss": 0.0025,
632
+ "step": 51000
633
+ },
634
+ {
635
+ "epoch": 1.4,
636
+ "learning_rate": 4.1274067796610175e-05,
637
+ "loss": 0.0032,
638
+ "step": 51500
639
+ },
640
+ {
641
+ "epoch": 1.41,
642
+ "learning_rate": 4.118949152542373e-05,
643
+ "loss": 0.0026,
644
+ "step": 52000
645
+ },
646
+ {
647
+ "epoch": 1.42,
648
+ "learning_rate": 4.1104745762711864e-05,
649
+ "loss": 0.0026,
650
+ "step": 52500
651
+ },
652
+ {
653
+ "epoch": 1.44,
654
+ "learning_rate": 4.1020000000000004e-05,
655
+ "loss": 0.0025,
656
+ "step": 53000
657
+ },
658
+ {
659
+ "epoch": 1.45,
660
+ "learning_rate": 4.093525423728814e-05,
661
+ "loss": 0.0028,
662
+ "step": 53500
663
+ },
664
+ {
665
+ "epoch": 1.46,
666
+ "learning_rate": 4.085050847457627e-05,
667
+ "loss": 0.0027,
668
+ "step": 54000
669
+ },
670
+ {
671
+ "epoch": 1.48,
672
+ "learning_rate": 4.076593220338983e-05,
673
+ "loss": 0.0027,
674
+ "step": 54500
675
+ },
676
+ {
677
+ "epoch": 1.49,
678
+ "learning_rate": 4.068118644067797e-05,
679
+ "loss": 0.0027,
680
+ "step": 55000
681
+ },
682
+ {
683
+ "epoch": 1.51,
684
+ "learning_rate": 4.0596440677966105e-05,
685
+ "loss": 0.0024,
686
+ "step": 55500
687
+ },
688
+ {
689
+ "epoch": 1.52,
690
+ "learning_rate": 4.051169491525424e-05,
691
+ "loss": 0.0026,
692
+ "step": 56000
693
+ },
694
+ {
695
+ "epoch": 1.53,
696
+ "learning_rate": 4.042694915254238e-05,
697
+ "loss": 0.0025,
698
+ "step": 56500
699
+ },
700
+ {
701
+ "epoch": 1.55,
702
+ "learning_rate": 4.034220338983051e-05,
703
+ "loss": 0.0026,
704
+ "step": 57000
705
+ },
706
+ {
707
+ "epoch": 1.56,
708
+ "learning_rate": 4.0257457627118645e-05,
709
+ "loss": 0.0023,
710
+ "step": 57500
711
+ },
712
+ {
713
+ "epoch": 1.57,
714
+ "learning_rate": 4.017271186440678e-05,
715
+ "loss": 0.0027,
716
+ "step": 58000
717
+ },
718
+ {
719
+ "epoch": 1.59,
720
+ "learning_rate": 4.008796610169492e-05,
721
+ "loss": 0.0025,
722
+ "step": 58500
723
+ },
724
+ {
725
+ "epoch": 1.6,
726
+ "learning_rate": 4.000322033898306e-05,
727
+ "loss": 0.0026,
728
+ "step": 59000
729
+ },
730
+ {
731
+ "epoch": 1.61,
732
+ "learning_rate": 3.9918474576271184e-05,
733
+ "loss": 0.0028,
734
+ "step": 59500
735
+ },
736
+ {
737
+ "epoch": 1.63,
738
+ "learning_rate": 3.9833728813559324e-05,
739
+ "loss": 0.0027,
740
+ "step": 60000
741
+ },
742
+ {
743
+ "epoch": 1.64,
744
+ "learning_rate": 3.974898305084746e-05,
745
+ "loss": 0.0026,
746
+ "step": 60500
747
+ },
748
+ {
749
+ "epoch": 1.65,
750
+ "learning_rate": 3.966440677966102e-05,
751
+ "loss": 0.0026,
752
+ "step": 61000
753
+ },
754
+ {
755
+ "epoch": 1.67,
756
+ "learning_rate": 3.957966101694915e-05,
757
+ "loss": 0.0026,
758
+ "step": 61500
759
+ },
760
+ {
761
+ "epoch": 1.68,
762
+ "learning_rate": 3.9495084745762715e-05,
763
+ "loss": 0.0026,
764
+ "step": 62000
765
+ },
766
+ {
767
+ "epoch": 1.69,
768
+ "learning_rate": 3.941033898305085e-05,
769
+ "loss": 0.0026,
770
+ "step": 62500
771
+ },
772
+ {
773
+ "epoch": 1.71,
774
+ "learning_rate": 3.932559322033899e-05,
775
+ "loss": 0.0028,
776
+ "step": 63000
777
+ },
778
+ {
779
+ "epoch": 1.72,
780
+ "learning_rate": 3.924084745762712e-05,
781
+ "loss": 0.0028,
782
+ "step": 63500
783
+ },
784
+ {
785
+ "epoch": 1.74,
786
+ "learning_rate": 3.9156101694915254e-05,
787
+ "loss": 0.0028,
788
+ "step": 64000
789
+ },
790
+ {
791
+ "epoch": 1.75,
792
+ "learning_rate": 3.907135593220339e-05,
793
+ "loss": 0.0028,
794
+ "step": 64500
795
+ },
796
+ {
797
+ "epoch": 1.76,
798
+ "learning_rate": 3.898661016949153e-05,
799
+ "loss": 0.0027,
800
+ "step": 65000
801
+ },
802
+ {
803
+ "epoch": 1.78,
804
+ "learning_rate": 3.890186440677967e-05,
805
+ "loss": 0.0027,
806
+ "step": 65500
807
+ },
808
+ {
809
+ "epoch": 1.79,
810
+ "learning_rate": 3.8817118644067794e-05,
811
+ "loss": 0.0027,
812
+ "step": 66000
813
+ },
814
+ {
815
+ "epoch": 1.8,
816
+ "learning_rate": 3.8732542372881356e-05,
817
+ "loss": 0.0024,
818
+ "step": 66500
819
+ },
820
+ {
821
+ "epoch": 1.82,
822
+ "learning_rate": 3.8647796610169496e-05,
823
+ "loss": 0.0028,
824
+ "step": 67000
825
+ },
826
+ {
827
+ "epoch": 1.83,
828
+ "learning_rate": 3.856305084745763e-05,
829
+ "loss": 0.0026,
830
+ "step": 67500
831
+ },
832
+ {
833
+ "epoch": 1.84,
834
+ "learning_rate": 3.847830508474576e-05,
835
+ "loss": 0.0028,
836
+ "step": 68000
837
+ },
838
+ {
839
+ "epoch": 1.86,
840
+ "learning_rate": 3.83935593220339e-05,
841
+ "loss": 0.0025,
842
+ "step": 68500
843
+ },
844
+ {
845
+ "epoch": 1.87,
846
+ "learning_rate": 3.830898305084746e-05,
847
+ "loss": 0.0026,
848
+ "step": 69000
849
+ },
850
+ {
851
+ "epoch": 1.88,
852
+ "learning_rate": 3.822457627118644e-05,
853
+ "loss": 0.0025,
854
+ "step": 69500
855
+ },
856
+ {
857
+ "epoch": 1.9,
858
+ "learning_rate": 3.8139830508474575e-05,
859
+ "loss": 0.0026,
860
+ "step": 70000
861
+ },
862
+ {
863
+ "epoch": 1.91,
864
+ "learning_rate": 3.8055084745762715e-05,
865
+ "loss": 0.0027,
866
+ "step": 70500
867
+ },
868
+ {
869
+ "epoch": 1.93,
870
+ "learning_rate": 3.797033898305085e-05,
871
+ "loss": 0.0027,
872
+ "step": 71000
873
+ },
874
+ {
875
+ "epoch": 1.94,
876
+ "learning_rate": 3.788559322033898e-05,
877
+ "loss": 0.0027,
878
+ "step": 71500
879
+ },
880
+ {
881
+ "epoch": 1.95,
882
+ "learning_rate": 3.780084745762712e-05,
883
+ "loss": 0.0026,
884
+ "step": 72000
885
+ },
886
+ {
887
+ "epoch": 1.97,
888
+ "learning_rate": 3.7716101694915255e-05,
889
+ "loss": 0.0026,
890
+ "step": 72500
891
+ },
892
+ {
893
+ "epoch": 1.98,
894
+ "learning_rate": 3.7631355932203395e-05,
895
+ "loss": 0.0022,
896
+ "step": 73000
897
+ },
898
+ {
899
+ "epoch": 1.99,
900
+ "learning_rate": 3.754677966101695e-05,
901
+ "loss": 0.0026,
902
+ "step": 73500
903
+ },
904
+ {
905
+ "epoch": 2.0,
906
+ "eval_accuracy": 0.9992288213058589,
907
+ "eval_f1": 0.0,
908
+ "eval_loss": 0.002819355810061097,
909
+ "eval_precision": 0.0,
910
+ "eval_recall": 0.0,
911
+ "eval_runtime": 364.7923,
912
+ "eval_samples_per_second": 108.58,
913
+ "eval_steps_per_second": 6.787,
914
+ "step": 73750
915
+ },
916
+ {
917
+ "epoch": 2.01,
918
+ "learning_rate": 3.746203389830509e-05,
919
+ "loss": 0.0021,
920
+ "step": 74000
921
+ },
922
+ {
923
+ "epoch": 2.02,
924
+ "learning_rate": 3.737728813559322e-05,
925
+ "loss": 0.0017,
926
+ "step": 74500
927
+ },
928
+ {
929
+ "epoch": 2.03,
930
+ "learning_rate": 3.729271186440678e-05,
931
+ "loss": 0.0016,
932
+ "step": 75000
933
+ },
934
+ {
935
+ "epoch": 2.05,
936
+ "learning_rate": 3.720796610169492e-05,
937
+ "loss": 0.0018,
938
+ "step": 75500
939
+ },
940
+ {
941
+ "epoch": 2.06,
942
+ "learning_rate": 3.712322033898305e-05,
943
+ "loss": 0.0016,
944
+ "step": 76000
945
+ },
946
+ {
947
+ "epoch": 2.07,
948
+ "learning_rate": 3.7038644067796614e-05,
949
+ "loss": 0.0017,
950
+ "step": 76500
951
+ },
952
+ {
953
+ "epoch": 2.09,
954
+ "learning_rate": 3.695389830508475e-05,
955
+ "loss": 0.0018,
956
+ "step": 77000
957
+ },
958
+ {
959
+ "epoch": 2.1,
960
+ "learning_rate": 3.686932203389831e-05,
961
+ "loss": 0.0022,
962
+ "step": 77500
963
+ },
964
+ {
965
+ "epoch": 2.12,
966
+ "learning_rate": 3.678457627118644e-05,
967
+ "loss": 0.0017,
968
+ "step": 78000
969
+ },
970
+ {
971
+ "epoch": 2.13,
972
+ "learning_rate": 3.6699830508474575e-05,
973
+ "loss": 0.0016,
974
+ "step": 78500
975
+ },
976
+ {
977
+ "epoch": 2.14,
978
+ "learning_rate": 3.6615084745762715e-05,
979
+ "loss": 0.0017,
980
+ "step": 79000
981
+ },
982
+ {
983
+ "epoch": 2.16,
984
+ "learning_rate": 3.653033898305085e-05,
985
+ "loss": 0.0018,
986
+ "step": 79500
987
+ },
988
+ {
989
+ "epoch": 2.17,
990
+ "learning_rate": 3.644559322033899e-05,
991
+ "loss": 0.0018,
992
+ "step": 80000
993
+ },
994
+ {
995
+ "epoch": 2.18,
996
+ "learning_rate": 3.636084745762712e-05,
997
+ "loss": 0.0017,
998
+ "step": 80500
999
+ },
1000
+ {
1001
+ "epoch": 2.2,
1002
+ "learning_rate": 3.627627118644068e-05,
1003
+ "loss": 0.0018,
1004
+ "step": 81000
1005
+ },
1006
+ {
1007
+ "epoch": 2.21,
1008
+ "learning_rate": 3.619152542372882e-05,
1009
+ "loss": 0.0016,
1010
+ "step": 81500
1011
+ },
1012
+ {
1013
+ "epoch": 2.22,
1014
+ "learning_rate": 3.610677966101695e-05,
1015
+ "loss": 0.0016,
1016
+ "step": 82000
1017
+ },
1018
+ {
1019
+ "epoch": 2.24,
1020
+ "learning_rate": 3.6022033898305083e-05,
1021
+ "loss": 0.0015,
1022
+ "step": 82500
1023
+ },
1024
+ {
1025
+ "epoch": 2.25,
1026
+ "learning_rate": 3.5937288135593223e-05,
1027
+ "loss": 0.0017,
1028
+ "step": 83000
1029
+ },
1030
+ {
1031
+ "epoch": 2.26,
1032
+ "learning_rate": 3.5852542372881357e-05,
1033
+ "loss": 0.0018,
1034
+ "step": 83500
1035
+ },
1036
+ {
1037
+ "epoch": 2.28,
1038
+ "learning_rate": 3.576779661016949e-05,
1039
+ "loss": 0.0015,
1040
+ "step": 84000
1041
+ },
1042
+ {
1043
+ "epoch": 2.29,
1044
+ "learning_rate": 3.568305084745763e-05,
1045
+ "loss": 0.0019,
1046
+ "step": 84500
1047
+ },
1048
+ {
1049
+ "epoch": 2.31,
1050
+ "learning_rate": 3.559830508474576e-05,
1051
+ "loss": 0.0017,
1052
+ "step": 85000
1053
+ },
1054
+ {
1055
+ "epoch": 2.32,
1056
+ "learning_rate": 3.55135593220339e-05,
1057
+ "loss": 0.0018,
1058
+ "step": 85500
1059
+ },
1060
+ {
1061
+ "epoch": 2.33,
1062
+ "learning_rate": 3.5428813559322036e-05,
1063
+ "loss": 0.0018,
1064
+ "step": 86000
1065
+ },
1066
+ {
1067
+ "epoch": 2.35,
1068
+ "learning_rate": 3.534406779661017e-05,
1069
+ "loss": 0.0018,
1070
+ "step": 86500
1071
+ },
1072
+ {
1073
+ "epoch": 2.36,
1074
+ "learning_rate": 3.525932203389831e-05,
1075
+ "loss": 0.0018,
1076
+ "step": 87000
1077
+ },
1078
+ {
1079
+ "epoch": 2.37,
1080
+ "learning_rate": 3.517457627118644e-05,
1081
+ "loss": 0.0017,
1082
+ "step": 87500
1083
+ },
1084
+ {
1085
+ "epoch": 2.39,
1086
+ "learning_rate": 3.508983050847458e-05,
1087
+ "loss": 0.0019,
1088
+ "step": 88000
1089
+ },
1090
+ {
1091
+ "epoch": 2.4,
1092
+ "learning_rate": 3.500508474576271e-05,
1093
+ "loss": 0.002,
1094
+ "step": 88500
1095
+ },
1096
+ {
1097
+ "epoch": 2.41,
1098
+ "learning_rate": 3.492033898305085e-05,
1099
+ "loss": 0.0016,
1100
+ "step": 89000
1101
+ },
1102
+ {
1103
+ "epoch": 2.43,
1104
+ "learning_rate": 3.483559322033899e-05,
1105
+ "loss": 0.0014,
1106
+ "step": 89500
1107
+ },
1108
+ {
1109
+ "epoch": 2.44,
1110
+ "learning_rate": 3.4751016949152544e-05,
1111
+ "loss": 0.0018,
1112
+ "step": 90000
1113
+ },
1114
+ {
1115
+ "epoch": 2.45,
1116
+ "learning_rate": 3.466627118644068e-05,
1117
+ "loss": 0.0019,
1118
+ "step": 90500
1119
+ },
1120
+ {
1121
+ "epoch": 2.47,
1122
+ "learning_rate": 3.458152542372882e-05,
1123
+ "loss": 0.0019,
1124
+ "step": 91000
1125
+ },
1126
+ {
1127
+ "epoch": 2.48,
1128
+ "learning_rate": 3.449677966101695e-05,
1129
+ "loss": 0.0018,
1130
+ "step": 91500
1131
+ },
1132
+ {
1133
+ "epoch": 2.49,
1134
+ "learning_rate": 3.4412033898305084e-05,
1135
+ "loss": 0.0021,
1136
+ "step": 92000
1137
+ },
1138
+ {
1139
+ "epoch": 2.51,
1140
+ "learning_rate": 3.4327288135593224e-05,
1141
+ "loss": 0.0019,
1142
+ "step": 92500
1143
+ },
1144
+ {
1145
+ "epoch": 2.52,
1146
+ "learning_rate": 3.424254237288136e-05,
1147
+ "loss": 0.0018,
1148
+ "step": 93000
1149
+ },
1150
+ {
1151
+ "epoch": 2.54,
1152
+ "learning_rate": 3.41577966101695e-05,
1153
+ "loss": 0.0019,
1154
+ "step": 93500
1155
+ },
1156
+ {
1157
+ "epoch": 2.55,
1158
+ "learning_rate": 3.407305084745763e-05,
1159
+ "loss": 0.0017,
1160
+ "step": 94000
1161
+ },
1162
+ {
1163
+ "epoch": 2.56,
1164
+ "learning_rate": 3.398864406779661e-05,
1165
+ "loss": 0.0016,
1166
+ "step": 94500
1167
+ },
1168
+ {
1169
+ "epoch": 2.58,
1170
+ "learning_rate": 3.390389830508475e-05,
1171
+ "loss": 0.0017,
1172
+ "step": 95000
1173
+ },
1174
+ {
1175
+ "epoch": 2.59,
1176
+ "learning_rate": 3.381915254237288e-05,
1177
+ "loss": 0.0015,
1178
+ "step": 95500
1179
+ },
1180
+ {
1181
+ "epoch": 2.6,
1182
+ "learning_rate": 3.373440677966102e-05,
1183
+ "loss": 0.0019,
1184
+ "step": 96000
1185
+ },
1186
+ {
1187
+ "epoch": 2.62,
1188
+ "learning_rate": 3.3649830508474576e-05,
1189
+ "loss": 0.0019,
1190
+ "step": 96500
1191
+ },
1192
+ {
1193
+ "epoch": 2.63,
1194
+ "learning_rate": 3.3565084745762716e-05,
1195
+ "loss": 0.0019,
1196
+ "step": 97000
1197
+ },
1198
+ {
1199
+ "epoch": 2.64,
1200
+ "learning_rate": 3.348033898305085e-05,
1201
+ "loss": 0.0019,
1202
+ "step": 97500
1203
+ },
1204
+ {
1205
+ "epoch": 2.66,
1206
+ "learning_rate": 3.339559322033898e-05,
1207
+ "loss": 0.002,
1208
+ "step": 98000
1209
+ },
1210
+ {
1211
+ "epoch": 2.67,
1212
+ "learning_rate": 3.3311016949152544e-05,
1213
+ "loss": 0.0019,
1214
+ "step": 98500
1215
+ },
1216
+ {
1217
+ "epoch": 2.68,
1218
+ "learning_rate": 3.322627118644068e-05,
1219
+ "loss": 0.0018,
1220
+ "step": 99000
1221
+ },
1222
+ {
1223
+ "epoch": 2.7,
1224
+ "learning_rate": 3.314152542372882e-05,
1225
+ "loss": 0.0016,
1226
+ "step": 99500
1227
+ },
1228
+ {
1229
+ "epoch": 2.71,
1230
+ "learning_rate": 3.305677966101695e-05,
1231
+ "loss": 0.0019,
1232
+ "step": 100000
1233
+ },
1234
+ {
1235
+ "epoch": 2.73,
1236
+ "learning_rate": 3.2972203389830506e-05,
1237
+ "loss": 0.0016,
1238
+ "step": 100500
1239
+ },
1240
+ {
1241
+ "epoch": 2.74,
1242
+ "learning_rate": 3.2887457627118646e-05,
1243
+ "loss": 0.0021,
1244
+ "step": 101000
1245
+ },
1246
+ {
1247
+ "epoch": 2.75,
1248
+ "learning_rate": 3.280271186440678e-05,
1249
+ "loss": 0.0017,
1250
+ "step": 101500
1251
+ },
1252
+ {
1253
+ "epoch": 2.77,
1254
+ "learning_rate": 3.271796610169492e-05,
1255
+ "loss": 0.0017,
1256
+ "step": 102000
1257
+ },
1258
+ {
1259
+ "epoch": 2.78,
1260
+ "learning_rate": 3.2633389830508474e-05,
1261
+ "loss": 0.0021,
1262
+ "step": 102500
1263
+ },
1264
+ {
1265
+ "epoch": 2.79,
1266
+ "learning_rate": 3.2548813559322037e-05,
1267
+ "loss": 0.0018,
1268
+ "step": 103000
1269
+ },
1270
+ {
1271
+ "epoch": 2.81,
1272
+ "learning_rate": 3.246406779661017e-05,
1273
+ "loss": 0.0017,
1274
+ "step": 103500
1275
+ },
1276
+ {
1277
+ "epoch": 2.82,
1278
+ "learning_rate": 3.237932203389831e-05,
1279
+ "loss": 0.0018,
1280
+ "step": 104000
1281
+ },
1282
+ {
1283
+ "epoch": 2.83,
1284
+ "learning_rate": 3.229457627118644e-05,
1285
+ "loss": 0.0017,
1286
+ "step": 104500
1287
+ },
1288
+ {
1289
+ "epoch": 2.85,
1290
+ "learning_rate": 3.2209830508474576e-05,
1291
+ "loss": 0.0018,
1292
+ "step": 105000
1293
+ },
1294
+ {
1295
+ "epoch": 2.86,
1296
+ "learning_rate": 3.212508474576271e-05,
1297
+ "loss": 0.0021,
1298
+ "step": 105500
1299
+ },
1300
+ {
1301
+ "epoch": 2.87,
1302
+ "learning_rate": 3.204033898305085e-05,
1303
+ "loss": 0.0017,
1304
+ "step": 106000
1305
+ },
1306
+ {
1307
+ "epoch": 2.89,
1308
+ "learning_rate": 3.195559322033899e-05,
1309
+ "loss": 0.0019,
1310
+ "step": 106500
1311
+ },
1312
+ {
1313
+ "epoch": 2.9,
1314
+ "learning_rate": 3.1870847457627116e-05,
1315
+ "loss": 0.0018,
1316
+ "step": 107000
1317
+ },
1318
+ {
1319
+ "epoch": 2.92,
1320
+ "learning_rate": 3.1786101694915256e-05,
1321
+ "loss": 0.0017,
1322
+ "step": 107500
1323
+ },
1324
+ {
1325
+ "epoch": 2.93,
1326
+ "learning_rate": 3.170135593220339e-05,
1327
+ "loss": 0.0019,
1328
+ "step": 108000
1329
+ },
1330
+ {
1331
+ "epoch": 2.94,
1332
+ "learning_rate": 3.161661016949153e-05,
1333
+ "loss": 0.0018,
1334
+ "step": 108500
1335
+ },
1336
+ {
1337
+ "epoch": 2.96,
1338
+ "learning_rate": 3.1532033898305084e-05,
1339
+ "loss": 0.0019,
1340
+ "step": 109000
1341
+ },
1342
+ {
1343
+ "epoch": 2.97,
1344
+ "learning_rate": 3.1447288135593224e-05,
1345
+ "loss": 0.0034,
1346
+ "step": 109500
1347
+ },
1348
+ {
1349
+ "epoch": 2.98,
1350
+ "learning_rate": 3.136254237288136e-05,
1351
+ "loss": 0.0019,
1352
+ "step": 110000
1353
+ },
1354
+ {
1355
+ "epoch": 3.0,
1356
+ "learning_rate": 3.127779661016949e-05,
1357
+ "loss": 0.0021,
1358
+ "step": 110500
1359
+ },
1360
+ {
1361
+ "epoch": 3.0,
1362
+ "eval_accuracy": 0.999242562907679,
1363
+ "eval_f1": 0.0,
1364
+ "eval_loss": 0.003269583685323596,
1365
+ "eval_precision": 0.0,
1366
+ "eval_recall": 0.0,
1367
+ "eval_runtime": 363.311,
1368
+ "eval_samples_per_second": 109.022,
1369
+ "eval_steps_per_second": 6.815,
1370
+ "step": 110625
1371
+ },
1372
+ {
1373
+ "epoch": 3.01,
1374
+ "learning_rate": 3.119322033898305e-05,
1375
+ "loss": 0.0013,
1376
+ "step": 111000
1377
+ },
1378
+ {
1379
+ "epoch": 3.02,
1380
+ "learning_rate": 3.110864406779661e-05,
1381
+ "loss": 0.0011,
1382
+ "step": 111500
1383
+ },
1384
+ {
1385
+ "epoch": 3.04,
1386
+ "learning_rate": 3.102389830508475e-05,
1387
+ "loss": 0.001,
1388
+ "step": 112000
1389
+ },
1390
+ {
1391
+ "epoch": 3.05,
1392
+ "learning_rate": 3.093915254237289e-05,
1393
+ "loss": 0.0011,
1394
+ "step": 112500
1395
+ },
1396
+ {
1397
+ "epoch": 3.06,
1398
+ "learning_rate": 3.0854406779661014e-05,
1399
+ "loss": 0.0013,
1400
+ "step": 113000
1401
+ },
1402
+ {
1403
+ "epoch": 3.08,
1404
+ "learning_rate": 3.0769661016949154e-05,
1405
+ "loss": 0.0012,
1406
+ "step": 113500
1407
+ },
1408
+ {
1409
+ "epoch": 3.09,
1410
+ "learning_rate": 3.068491525423729e-05,
1411
+ "loss": 0.0013,
1412
+ "step": 114000
1413
+ },
1414
+ {
1415
+ "epoch": 3.11,
1416
+ "learning_rate": 3.060016949152543e-05,
1417
+ "loss": 0.0013,
1418
+ "step": 114500
1419
+ },
1420
+ {
1421
+ "epoch": 3.12,
1422
+ "learning_rate": 3.0515423728813557e-05,
1423
+ "loss": 0.0012,
1424
+ "step": 115000
1425
+ },
1426
+ {
1427
+ "epoch": 3.13,
1428
+ "learning_rate": 3.043084745762712e-05,
1429
+ "loss": 0.0015,
1430
+ "step": 115500
1431
+ },
1432
+ {
1433
+ "epoch": 3.15,
1434
+ "learning_rate": 3.034610169491526e-05,
1435
+ "loss": 0.0013,
1436
+ "step": 116000
1437
+ },
1438
+ {
1439
+ "epoch": 3.16,
1440
+ "learning_rate": 3.026135593220339e-05,
1441
+ "loss": 0.0013,
1442
+ "step": 116500
1443
+ },
1444
+ {
1445
+ "epoch": 3.17,
1446
+ "learning_rate": 3.0176610169491526e-05,
1447
+ "loss": 0.0015,
1448
+ "step": 117000
1449
+ },
1450
+ {
1451
+ "epoch": 3.19,
1452
+ "learning_rate": 3.0091864406779662e-05,
1453
+ "loss": 0.0014,
1454
+ "step": 117500
1455
+ },
1456
+ {
1457
+ "epoch": 3.2,
1458
+ "learning_rate": 3.000728813559322e-05,
1459
+ "loss": 0.0015,
1460
+ "step": 118000
1461
+ },
1462
+ {
1463
+ "epoch": 3.21,
1464
+ "learning_rate": 2.9922711864406783e-05,
1465
+ "loss": 0.0013,
1466
+ "step": 118500
1467
+ },
1468
+ {
1469
+ "epoch": 3.23,
1470
+ "learning_rate": 2.9837966101694913e-05,
1471
+ "loss": 0.0012,
1472
+ "step": 119000
1473
+ },
1474
+ {
1475
+ "epoch": 3.24,
1476
+ "learning_rate": 2.975322033898305e-05,
1477
+ "loss": 0.0012,
1478
+ "step": 119500
1479
+ },
1480
+ {
1481
+ "epoch": 3.25,
1482
+ "learning_rate": 2.966847457627119e-05,
1483
+ "loss": 0.0015,
1484
+ "step": 120000
1485
+ },
1486
+ {
1487
+ "epoch": 3.27,
1488
+ "learning_rate": 2.9583728813559326e-05,
1489
+ "loss": 0.0014,
1490
+ "step": 120500
1491
+ },
1492
+ {
1493
+ "epoch": 3.28,
1494
+ "learning_rate": 2.9498983050847456e-05,
1495
+ "loss": 0.0012,
1496
+ "step": 121000
1497
+ },
1498
+ {
1499
+ "epoch": 3.29,
1500
+ "learning_rate": 2.9414237288135592e-05,
1501
+ "loss": 0.0011,
1502
+ "step": 121500
1503
+ },
1504
+ {
1505
+ "epoch": 3.31,
1506
+ "learning_rate": 2.932949152542373e-05,
1507
+ "loss": 0.0012,
1508
+ "step": 122000
1509
+ },
1510
+ {
1511
+ "epoch": 3.32,
1512
+ "learning_rate": 2.9244745762711865e-05,
1513
+ "loss": 0.0013,
1514
+ "step": 122500
1515
+ },
1516
+ {
1517
+ "epoch": 3.34,
1518
+ "learning_rate": 2.9160000000000005e-05,
1519
+ "loss": 0.0014,
1520
+ "step": 123000
1521
+ },
1522
+ {
1523
+ "epoch": 3.35,
1524
+ "learning_rate": 2.9075254237288135e-05,
1525
+ "loss": 0.0012,
1526
+ "step": 123500
1527
+ },
1528
+ {
1529
+ "epoch": 3.36,
1530
+ "learning_rate": 2.8990508474576272e-05,
1531
+ "loss": 0.0014,
1532
+ "step": 124000
1533
+ },
1534
+ {
1535
+ "epoch": 3.38,
1536
+ "learning_rate": 2.890593220338983e-05,
1537
+ "loss": 0.0013,
1538
+ "step": 124500
1539
+ },
1540
+ {
1541
+ "epoch": 3.39,
1542
+ "learning_rate": 2.8821355932203393e-05,
1543
+ "loss": 0.0016,
1544
+ "step": 125000
1545
+ },
1546
+ {
1547
+ "epoch": 3.4,
1548
+ "learning_rate": 2.873661016949153e-05,
1549
+ "loss": 0.0012,
1550
+ "step": 125500
1551
+ },
1552
+ {
1553
+ "epoch": 3.42,
1554
+ "learning_rate": 2.865186440677966e-05,
1555
+ "loss": 0.0012,
1556
+ "step": 126000
1557
+ },
1558
+ {
1559
+ "epoch": 3.43,
1560
+ "learning_rate": 2.85671186440678e-05,
1561
+ "loss": 0.0013,
1562
+ "step": 126500
1563
+ },
1564
+ {
1565
+ "epoch": 3.44,
1566
+ "learning_rate": 2.8482372881355936e-05,
1567
+ "loss": 0.0012,
1568
+ "step": 127000
1569
+ },
1570
+ {
1571
+ "epoch": 3.46,
1572
+ "learning_rate": 2.8397627118644072e-05,
1573
+ "loss": 0.0012,
1574
+ "step": 127500
1575
+ },
1576
+ {
1577
+ "epoch": 3.47,
1578
+ "learning_rate": 2.8312881355932202e-05,
1579
+ "loss": 0.0013,
1580
+ "step": 128000
1581
+ },
1582
+ {
1583
+ "epoch": 3.48,
1584
+ "learning_rate": 2.822813559322034e-05,
1585
+ "loss": 0.0011,
1586
+ "step": 128500
1587
+ },
1588
+ {
1589
+ "epoch": 3.5,
1590
+ "learning_rate": 2.81435593220339e-05,
1591
+ "loss": 0.0014,
1592
+ "step": 129000
1593
+ },
1594
+ {
1595
+ "epoch": 3.51,
1596
+ "learning_rate": 2.8058813559322034e-05,
1597
+ "loss": 0.0015,
1598
+ "step": 129500
1599
+ },
1600
+ {
1601
+ "epoch": 3.53,
1602
+ "learning_rate": 2.797406779661017e-05,
1603
+ "loss": 0.0012,
1604
+ "step": 130000
1605
+ },
1606
+ {
1607
+ "epoch": 3.54,
1608
+ "learning_rate": 2.7889322033898307e-05,
1609
+ "loss": 0.0012,
1610
+ "step": 130500
1611
+ },
1612
+ {
1613
+ "epoch": 3.55,
1614
+ "learning_rate": 2.7804576271186444e-05,
1615
+ "loss": 0.0013,
1616
+ "step": 131000
1617
+ },
1618
+ {
1619
+ "epoch": 3.57,
1620
+ "learning_rate": 2.7719830508474577e-05,
1621
+ "loss": 0.0011,
1622
+ "step": 131500
1623
+ },
1624
+ {
1625
+ "epoch": 3.58,
1626
+ "learning_rate": 2.7635084745762713e-05,
1627
+ "loss": 0.0013,
1628
+ "step": 132000
1629
+ },
1630
+ {
1631
+ "epoch": 3.59,
1632
+ "learning_rate": 2.755033898305085e-05,
1633
+ "loss": 0.0012,
1634
+ "step": 132500
1635
+ },
1636
+ {
1637
+ "epoch": 3.61,
1638
+ "learning_rate": 2.7465593220338986e-05,
1639
+ "loss": 0.0013,
1640
+ "step": 133000
1641
+ },
1642
+ {
1643
+ "epoch": 3.62,
1644
+ "learning_rate": 2.7380847457627116e-05,
1645
+ "loss": 0.0012,
1646
+ "step": 133500
1647
+ },
1648
+ {
1649
+ "epoch": 3.63,
1650
+ "learning_rate": 2.7296271186440682e-05,
1651
+ "loss": 0.0013,
1652
+ "step": 134000
1653
+ },
1654
+ {
1655
+ "epoch": 3.65,
1656
+ "learning_rate": 2.7211694915254237e-05,
1657
+ "loss": 0.0015,
1658
+ "step": 134500
1659
+ },
1660
+ {
1661
+ "epoch": 3.66,
1662
+ "learning_rate": 2.7126949152542374e-05,
1663
+ "loss": 0.0012,
1664
+ "step": 135000
1665
+ },
1666
+ {
1667
+ "epoch": 3.67,
1668
+ "learning_rate": 2.704220338983051e-05,
1669
+ "loss": 0.0013,
1670
+ "step": 135500
1671
+ },
1672
+ {
1673
+ "epoch": 3.69,
1674
+ "learning_rate": 2.6957457627118643e-05,
1675
+ "loss": 0.0013,
1676
+ "step": 136000
1677
+ },
1678
+ {
1679
+ "epoch": 3.7,
1680
+ "learning_rate": 2.687271186440678e-05,
1681
+ "loss": 0.0013,
1682
+ "step": 136500
1683
+ },
1684
+ {
1685
+ "epoch": 3.72,
1686
+ "learning_rate": 2.6787966101694917e-05,
1687
+ "loss": 0.0012,
1688
+ "step": 137000
1689
+ },
1690
+ {
1691
+ "epoch": 3.73,
1692
+ "learning_rate": 2.6703220338983053e-05,
1693
+ "loss": 0.0014,
1694
+ "step": 137500
1695
+ },
1696
+ {
1697
+ "epoch": 3.74,
1698
+ "learning_rate": 2.6618644067796612e-05,
1699
+ "loss": 0.0012,
1700
+ "step": 138000
1701
+ },
1702
+ {
1703
+ "epoch": 3.76,
1704
+ "learning_rate": 2.6534067796610167e-05,
1705
+ "loss": 0.0012,
1706
+ "step": 138500
1707
+ },
1708
+ {
1709
+ "epoch": 3.77,
1710
+ "learning_rate": 2.6449322033898304e-05,
1711
+ "loss": 0.0013,
1712
+ "step": 139000
1713
+ },
1714
+ {
1715
+ "epoch": 3.78,
1716
+ "learning_rate": 2.636457627118644e-05,
1717
+ "loss": 0.0012,
1718
+ "step": 139500
1719
+ },
1720
+ {
1721
+ "epoch": 3.8,
1722
+ "learning_rate": 2.627983050847458e-05,
1723
+ "loss": 0.0012,
1724
+ "step": 140000
1725
+ },
1726
+ {
1727
+ "epoch": 3.81,
1728
+ "learning_rate": 2.6195084745762717e-05,
1729
+ "loss": 0.0013,
1730
+ "step": 140500
1731
+ },
1732
+ {
1733
+ "epoch": 3.82,
1734
+ "learning_rate": 2.6110338983050847e-05,
1735
+ "loss": 0.0013,
1736
+ "step": 141000
1737
+ },
1738
+ {
1739
+ "epoch": 3.84,
1740
+ "learning_rate": 2.6025593220338983e-05,
1741
+ "loss": 0.0011,
1742
+ "step": 141500
1743
+ },
1744
+ {
1745
+ "epoch": 3.85,
1746
+ "learning_rate": 2.594084745762712e-05,
1747
+ "loss": 0.0012,
1748
+ "step": 142000
1749
+ },
1750
+ {
1751
+ "epoch": 3.86,
1752
+ "learning_rate": 2.5856101694915256e-05,
1753
+ "loss": 0.0012,
1754
+ "step": 142500
1755
+ },
1756
+ {
1757
+ "epoch": 3.88,
1758
+ "learning_rate": 2.577135593220339e-05,
1759
+ "loss": 0.0015,
1760
+ "step": 143000
1761
+ },
1762
+ {
1763
+ "epoch": 3.89,
1764
+ "learning_rate": 2.5686610169491526e-05,
1765
+ "loss": 0.0012,
1766
+ "step": 143500
1767
+ },
1768
+ {
1769
+ "epoch": 3.91,
1770
+ "learning_rate": 2.5601864406779663e-05,
1771
+ "loss": 0.0012,
1772
+ "step": 144000
1773
+ },
1774
+ {
1775
+ "epoch": 3.92,
1776
+ "learning_rate": 2.551728813559322e-05,
1777
+ "loss": 0.0011,
1778
+ "step": 144500
1779
+ },
1780
+ {
1781
+ "epoch": 3.93,
1782
+ "learning_rate": 2.5432542372881358e-05,
1783
+ "loss": 0.0011,
1784
+ "step": 145000
1785
+ },
1786
+ {
1787
+ "epoch": 3.95,
1788
+ "learning_rate": 2.5347796610169495e-05,
1789
+ "loss": 0.0013,
1790
+ "step": 145500
1791
+ },
1792
+ {
1793
+ "epoch": 3.96,
1794
+ "learning_rate": 2.526322033898305e-05,
1795
+ "loss": 0.0013,
1796
+ "step": 146000
1797
+ },
1798
+ {
1799
+ "epoch": 3.97,
1800
+ "learning_rate": 2.5178474576271187e-05,
1801
+ "loss": 0.0012,
1802
+ "step": 146500
1803
+ },
1804
+ {
1805
+ "epoch": 3.99,
1806
+ "learning_rate": 2.5093728813559327e-05,
1807
+ "loss": 0.0011,
1808
+ "step": 147000
1809
+ },
1810
+ {
1811
+ "epoch": 4.0,
1812
+ "learning_rate": 2.5008983050847456e-05,
1813
+ "loss": 0.0014,
1814
+ "step": 147500
1815
+ },
1816
+ {
1817
+ "epoch": 4.0,
1818
+ "eval_accuracy": 0.9992986286431048,
1819
+ "eval_f1": 0.0,
1820
+ "eval_loss": 0.003328560618683696,
1821
+ "eval_precision": 0.0,
1822
+ "eval_recall": 0.0,
1823
+ "eval_runtime": 365.3105,
1824
+ "eval_samples_per_second": 108.426,
1825
+ "eval_steps_per_second": 6.778,
1826
+ "step": 147500
1827
+ },
1828
+ {
1829
+ "epoch": 4.01,
1830
+ "learning_rate": 2.4924237288135593e-05,
1831
+ "loss": 0.0008,
1832
+ "step": 148000
1833
+ },
1834
+ {
1835
+ "epoch": 4.03,
1836
+ "learning_rate": 2.483949152542373e-05,
1837
+ "loss": 0.0009,
1838
+ "step": 148500
1839
+ },
1840
+ {
1841
+ "epoch": 4.04,
1842
+ "learning_rate": 2.4754745762711866e-05,
1843
+ "loss": 0.0009,
1844
+ "step": 149000
1845
+ },
1846
+ {
1847
+ "epoch": 4.05,
1848
+ "learning_rate": 2.4670000000000003e-05,
1849
+ "loss": 0.0007,
1850
+ "step": 149500
1851
+ },
1852
+ {
1853
+ "epoch": 4.07,
1854
+ "learning_rate": 2.458542372881356e-05,
1855
+ "loss": 0.0009,
1856
+ "step": 150000
1857
+ },
1858
+ {
1859
+ "epoch": 4.08,
1860
+ "learning_rate": 2.4500677966101695e-05,
1861
+ "loss": 0.0008,
1862
+ "step": 150500
1863
+ },
1864
+ {
1865
+ "epoch": 4.09,
1866
+ "learning_rate": 2.441593220338983e-05,
1867
+ "loss": 0.0006,
1868
+ "step": 151000
1869
+ },
1870
+ {
1871
+ "epoch": 4.11,
1872
+ "learning_rate": 2.4331186440677968e-05,
1873
+ "loss": 0.0009,
1874
+ "step": 151500
1875
+ },
1876
+ {
1877
+ "epoch": 4.12,
1878
+ "learning_rate": 2.4246440677966104e-05,
1879
+ "loss": 0.0007,
1880
+ "step": 152000
1881
+ },
1882
+ {
1883
+ "epoch": 4.14,
1884
+ "learning_rate": 2.4161694915254237e-05,
1885
+ "loss": 0.0007,
1886
+ "step": 152500
1887
+ },
1888
+ {
1889
+ "epoch": 4.15,
1890
+ "learning_rate": 2.4077288135593222e-05,
1891
+ "loss": 0.0009,
1892
+ "step": 153000
1893
+ },
1894
+ {
1895
+ "epoch": 4.16,
1896
+ "learning_rate": 2.399254237288136e-05,
1897
+ "loss": 0.0008,
1898
+ "step": 153500
1899
+ },
1900
+ {
1901
+ "epoch": 4.18,
1902
+ "learning_rate": 2.3907966101694917e-05,
1903
+ "loss": 0.0007,
1904
+ "step": 154000
1905
+ },
1906
+ {
1907
+ "epoch": 4.19,
1908
+ "learning_rate": 2.382322033898305e-05,
1909
+ "loss": 0.0009,
1910
+ "step": 154500
1911
+ },
1912
+ {
1913
+ "epoch": 4.2,
1914
+ "learning_rate": 2.373847457627119e-05,
1915
+ "loss": 0.0008,
1916
+ "step": 155000
1917
+ },
1918
+ {
1919
+ "epoch": 4.22,
1920
+ "learning_rate": 2.3653728813559323e-05,
1921
+ "loss": 0.001,
1922
+ "step": 155500
1923
+ },
1924
+ {
1925
+ "epoch": 4.23,
1926
+ "learning_rate": 2.356898305084746e-05,
1927
+ "loss": 0.001,
1928
+ "step": 156000
1929
+ },
1930
+ {
1931
+ "epoch": 4.24,
1932
+ "learning_rate": 2.3484237288135593e-05,
1933
+ "loss": 0.0009,
1934
+ "step": 156500
1935
+ },
1936
+ {
1937
+ "epoch": 4.26,
1938
+ "learning_rate": 2.339949152542373e-05,
1939
+ "loss": 0.0007,
1940
+ "step": 157000
1941
+ },
1942
+ {
1943
+ "epoch": 4.27,
1944
+ "learning_rate": 2.3314745762711866e-05,
1945
+ "loss": 0.001,
1946
+ "step": 157500
1947
+ },
1948
+ {
1949
+ "epoch": 4.28,
1950
+ "learning_rate": 2.3230000000000003e-05,
1951
+ "loss": 0.0008,
1952
+ "step": 158000
1953
+ },
1954
+ {
1955
+ "epoch": 4.3,
1956
+ "learning_rate": 2.3145254237288136e-05,
1957
+ "loss": 0.0006,
1958
+ "step": 158500
1959
+ },
1960
+ {
1961
+ "epoch": 4.31,
1962
+ "learning_rate": 2.3060677966101695e-05,
1963
+ "loss": 0.0009,
1964
+ "step": 159000
1965
+ },
1966
+ {
1967
+ "epoch": 4.33,
1968
+ "learning_rate": 2.297593220338983e-05,
1969
+ "loss": 0.001,
1970
+ "step": 159500
1971
+ },
1972
+ {
1973
+ "epoch": 4.34,
1974
+ "learning_rate": 2.2891186440677968e-05,
1975
+ "loss": 0.0007,
1976
+ "step": 160000
1977
+ },
1978
+ {
1979
+ "epoch": 4.35,
1980
+ "learning_rate": 2.2806440677966105e-05,
1981
+ "loss": 0.001,
1982
+ "step": 160500
1983
+ },
1984
+ {
1985
+ "epoch": 4.37,
1986
+ "learning_rate": 2.2721694915254238e-05,
1987
+ "loss": 0.001,
1988
+ "step": 161000
1989
+ },
1990
+ {
1991
+ "epoch": 4.38,
1992
+ "learning_rate": 2.2637118644067796e-05,
1993
+ "loss": 0.0008,
1994
+ "step": 161500
1995
+ },
1996
+ {
1997
+ "epoch": 4.39,
1998
+ "learning_rate": 2.2552372881355933e-05,
1999
+ "loss": 0.0008,
2000
+ "step": 162000
2001
+ },
2002
+ {
2003
+ "epoch": 4.41,
2004
+ "learning_rate": 2.246762711864407e-05,
2005
+ "loss": 0.0009,
2006
+ "step": 162500
2007
+ },
2008
+ {
2009
+ "epoch": 4.42,
2010
+ "learning_rate": 2.2382881355932203e-05,
2011
+ "loss": 0.001,
2012
+ "step": 163000
2013
+ },
2014
+ {
2015
+ "epoch": 4.43,
2016
+ "learning_rate": 2.229813559322034e-05,
2017
+ "loss": 0.0008,
2018
+ "step": 163500
2019
+ },
2020
+ {
2021
+ "epoch": 4.45,
2022
+ "learning_rate": 2.2213389830508476e-05,
2023
+ "loss": 0.0009,
2024
+ "step": 164000
2025
+ },
2026
+ {
2027
+ "epoch": 4.46,
2028
+ "learning_rate": 2.2128644067796613e-05,
2029
+ "loss": 0.001,
2030
+ "step": 164500
2031
+ },
2032
+ {
2033
+ "epoch": 4.47,
2034
+ "learning_rate": 2.204389830508475e-05,
2035
+ "loss": 0.0008,
2036
+ "step": 165000
2037
+ },
2038
+ {
2039
+ "epoch": 4.49,
2040
+ "learning_rate": 2.1959152542372882e-05,
2041
+ "loss": 0.0008,
2042
+ "step": 165500
2043
+ },
2044
+ {
2045
+ "epoch": 4.5,
2046
+ "learning_rate": 2.187457627118644e-05,
2047
+ "loss": 0.0009,
2048
+ "step": 166000
2049
+ },
2050
+ {
2051
+ "epoch": 4.52,
2052
+ "learning_rate": 2.1789830508474578e-05,
2053
+ "loss": 0.0008,
2054
+ "step": 166500
2055
+ },
2056
+ {
2057
+ "epoch": 4.53,
2058
+ "learning_rate": 2.1705084745762714e-05,
2059
+ "loss": 0.0009,
2060
+ "step": 167000
2061
+ },
2062
+ {
2063
+ "epoch": 4.54,
2064
+ "learning_rate": 2.1620508474576273e-05,
2065
+ "loss": 0.0009,
2066
+ "step": 167500
2067
+ },
2068
+ {
2069
+ "epoch": 4.56,
2070
+ "learning_rate": 2.1535762711864406e-05,
2071
+ "loss": 0.0007,
2072
+ "step": 168000
2073
+ },
2074
+ {
2075
+ "epoch": 4.57,
2076
+ "learning_rate": 2.1451016949152543e-05,
2077
+ "loss": 0.0009,
2078
+ "step": 168500
2079
+ },
2080
+ {
2081
+ "epoch": 4.58,
2082
+ "learning_rate": 2.136627118644068e-05,
2083
+ "loss": 0.0008,
2084
+ "step": 169000
2085
+ },
2086
+ {
2087
+ "epoch": 4.6,
2088
+ "learning_rate": 2.1281525423728816e-05,
2089
+ "loss": 0.0008,
2090
+ "step": 169500
2091
+ },
2092
+ {
2093
+ "epoch": 4.61,
2094
+ "learning_rate": 2.119677966101695e-05,
2095
+ "loss": 0.0008,
2096
+ "step": 170000
2097
+ },
2098
+ {
2099
+ "epoch": 4.62,
2100
+ "learning_rate": 2.1112203389830508e-05,
2101
+ "loss": 0.0008,
2102
+ "step": 170500
2103
+ },
2104
+ {
2105
+ "epoch": 4.64,
2106
+ "learning_rate": 2.1027457627118648e-05,
2107
+ "loss": 0.001,
2108
+ "step": 171000
2109
+ },
2110
+ {
2111
+ "epoch": 4.65,
2112
+ "learning_rate": 2.094271186440678e-05,
2113
+ "loss": 0.001,
2114
+ "step": 171500
2115
+ },
2116
+ {
2117
+ "epoch": 4.66,
2118
+ "learning_rate": 2.0857966101694917e-05,
2119
+ "loss": 0.0008,
2120
+ "step": 172000
2121
+ },
2122
+ {
2123
+ "epoch": 4.68,
2124
+ "learning_rate": 2.077322033898305e-05,
2125
+ "loss": 0.0009,
2126
+ "step": 172500
2127
+ },
2128
+ {
2129
+ "epoch": 4.69,
2130
+ "learning_rate": 2.0688474576271187e-05,
2131
+ "loss": 0.0008,
2132
+ "step": 173000
2133
+ },
2134
+ {
2135
+ "epoch": 4.71,
2136
+ "learning_rate": 2.0603728813559324e-05,
2137
+ "loss": 0.0009,
2138
+ "step": 173500
2139
+ },
2140
+ {
2141
+ "epoch": 4.72,
2142
+ "learning_rate": 2.051898305084746e-05,
2143
+ "loss": 0.001,
2144
+ "step": 174000
2145
+ },
2146
+ {
2147
+ "epoch": 4.73,
2148
+ "learning_rate": 2.0434406779661016e-05,
2149
+ "loss": 0.0008,
2150
+ "step": 174500
2151
+ },
2152
+ {
2153
+ "epoch": 4.75,
2154
+ "learning_rate": 2.0349661016949152e-05,
2155
+ "loss": 0.0009,
2156
+ "step": 175000
2157
+ },
2158
+ {
2159
+ "epoch": 4.76,
2160
+ "learning_rate": 2.026491525423729e-05,
2161
+ "loss": 0.0008,
2162
+ "step": 175500
2163
+ },
2164
+ {
2165
+ "epoch": 4.77,
2166
+ "learning_rate": 2.0180169491525425e-05,
2167
+ "loss": 0.0009,
2168
+ "step": 176000
2169
+ },
2170
+ {
2171
+ "epoch": 4.79,
2172
+ "learning_rate": 2.0095593220338984e-05,
2173
+ "loss": 0.0009,
2174
+ "step": 176500
2175
+ },
2176
+ {
2177
+ "epoch": 4.8,
2178
+ "learning_rate": 2.0010847457627117e-05,
2179
+ "loss": 0.0008,
2180
+ "step": 177000
2181
+ },
2182
+ {
2183
+ "epoch": 4.81,
2184
+ "learning_rate": 1.9926101694915257e-05,
2185
+ "loss": 0.0008,
2186
+ "step": 177500
2187
+ },
2188
+ {
2189
+ "epoch": 4.83,
2190
+ "learning_rate": 1.984135593220339e-05,
2191
+ "loss": 0.0007,
2192
+ "step": 178000
2193
+ },
2194
+ {
2195
+ "epoch": 4.84,
2196
+ "learning_rate": 1.9756610169491527e-05,
2197
+ "loss": 0.0007,
2198
+ "step": 178500
2199
+ },
2200
+ {
2201
+ "epoch": 4.85,
2202
+ "learning_rate": 1.967186440677966e-05,
2203
+ "loss": 0.0009,
2204
+ "step": 179000
2205
+ },
2206
+ {
2207
+ "epoch": 4.87,
2208
+ "learning_rate": 1.9587118644067797e-05,
2209
+ "loss": 0.0009,
2210
+ "step": 179500
2211
+ },
2212
+ {
2213
+ "epoch": 4.88,
2214
+ "learning_rate": 1.950254237288136e-05,
2215
+ "loss": 0.0008,
2216
+ "step": 180000
2217
+ },
2218
+ {
2219
+ "epoch": 4.89,
2220
+ "learning_rate": 1.9417966101694914e-05,
2221
+ "loss": 0.001,
2222
+ "step": 180500
2223
+ },
2224
+ {
2225
+ "epoch": 4.91,
2226
+ "learning_rate": 1.933322033898305e-05,
2227
+ "loss": 0.0008,
2228
+ "step": 181000
2229
+ },
2230
+ {
2231
+ "epoch": 4.92,
2232
+ "learning_rate": 1.9248474576271187e-05,
2233
+ "loss": 0.0008,
2234
+ "step": 181500
2235
+ },
2236
+ {
2237
+ "epoch": 4.94,
2238
+ "learning_rate": 1.9163728813559324e-05,
2239
+ "loss": 0.0008,
2240
+ "step": 182000
2241
+ },
2242
+ {
2243
+ "epoch": 4.95,
2244
+ "learning_rate": 1.907898305084746e-05,
2245
+ "loss": 0.0009,
2246
+ "step": 182500
2247
+ },
2248
+ {
2249
+ "epoch": 4.96,
2250
+ "learning_rate": 1.8994237288135594e-05,
2251
+ "loss": 0.0008,
2252
+ "step": 183000
2253
+ },
2254
+ {
2255
+ "epoch": 4.98,
2256
+ "learning_rate": 1.890949152542373e-05,
2257
+ "loss": 0.0007,
2258
+ "step": 183500
2259
+ },
2260
+ {
2261
+ "epoch": 4.99,
2262
+ "learning_rate": 1.8824745762711864e-05,
2263
+ "loss": 0.0009,
2264
+ "step": 184000
2265
+ },
2266
+ {
2267
+ "epoch": 5.0,
2268
+ "eval_accuracy": 0.9993343568078369,
2269
+ "eval_f1": 0.0,
2270
+ "eval_loss": 0.003261522389948368,
2271
+ "eval_precision": 0.0,
2272
+ "eval_recall": 0.0,
2273
+ "eval_runtime": 364.1478,
2274
+ "eval_samples_per_second": 108.772,
2275
+ "eval_steps_per_second": 6.799,
2276
+ "step": 184375
2277
+ },
2278
+ {
2279
+ "epoch": 5.0,
2280
+ "learning_rate": 1.8740000000000004e-05,
2281
+ "loss": 0.0008,
2282
+ "step": 184500
2283
+ },
2284
+ {
2285
+ "epoch": 5.02,
2286
+ "learning_rate": 1.865542372881356e-05,
2287
+ "loss": 0.0006,
2288
+ "step": 185000
2289
+ },
2290
+ {
2291
+ "epoch": 5.03,
2292
+ "learning_rate": 1.8570677966101695e-05,
2293
+ "loss": 0.0005,
2294
+ "step": 185500
2295
+ },
2296
+ {
2297
+ "epoch": 5.04,
2298
+ "learning_rate": 1.848593220338983e-05,
2299
+ "loss": 0.0005,
2300
+ "step": 186000
2301
+ },
2302
+ {
2303
+ "epoch": 5.06,
2304
+ "learning_rate": 1.840118644067797e-05,
2305
+ "loss": 0.0005,
2306
+ "step": 186500
2307
+ },
2308
+ {
2309
+ "epoch": 5.07,
2310
+ "learning_rate": 1.8316440677966102e-05,
2311
+ "loss": 0.0005,
2312
+ "step": 187000
2313
+ },
2314
+ {
2315
+ "epoch": 5.08,
2316
+ "learning_rate": 1.823169491525424e-05,
2317
+ "loss": 0.0006,
2318
+ "step": 187500
2319
+ },
2320
+ {
2321
+ "epoch": 5.1,
2322
+ "learning_rate": 1.814694915254237e-05,
2323
+ "loss": 0.0006,
2324
+ "step": 188000
2325
+ },
2326
+ {
2327
+ "epoch": 5.11,
2328
+ "learning_rate": 1.8062203389830508e-05,
2329
+ "loss": 0.0006,
2330
+ "step": 188500
2331
+ },
2332
+ {
2333
+ "epoch": 5.13,
2334
+ "learning_rate": 1.7977457627118645e-05,
2335
+ "loss": 0.0006,
2336
+ "step": 189000
2337
+ },
2338
+ {
2339
+ "epoch": 5.14,
2340
+ "learning_rate": 1.789271186440678e-05,
2341
+ "loss": 0.0005,
2342
+ "step": 189500
2343
+ },
2344
+ {
2345
+ "epoch": 5.15,
2346
+ "learning_rate": 1.780813559322034e-05,
2347
+ "loss": 0.0006,
2348
+ "step": 190000
2349
+ },
2350
+ {
2351
+ "epoch": 5.17,
2352
+ "learning_rate": 1.7723389830508473e-05,
2353
+ "loss": 0.0006,
2354
+ "step": 190500
2355
+ },
2356
+ {
2357
+ "epoch": 5.18,
2358
+ "learning_rate": 1.763864406779661e-05,
2359
+ "loss": 0.0006,
2360
+ "step": 191000
2361
+ },
2362
+ {
2363
+ "epoch": 5.19,
2364
+ "learning_rate": 1.7553898305084746e-05,
2365
+ "loss": 0.0006,
2366
+ "step": 191500
2367
+ },
2368
+ {
2369
+ "epoch": 5.21,
2370
+ "learning_rate": 1.7469152542372883e-05,
2371
+ "loss": 0.0006,
2372
+ "step": 192000
2373
+ },
2374
+ {
2375
+ "epoch": 5.22,
2376
+ "learning_rate": 1.7384406779661016e-05,
2377
+ "loss": 0.0006,
2378
+ "step": 192500
2379
+ },
2380
+ {
2381
+ "epoch": 5.23,
2382
+ "learning_rate": 1.7299661016949153e-05,
2383
+ "loss": 0.0006,
2384
+ "step": 193000
2385
+ },
2386
+ {
2387
+ "epoch": 5.25,
2388
+ "learning_rate": 1.721491525423729e-05,
2389
+ "loss": 0.0004,
2390
+ "step": 193500
2391
+ },
2392
+ {
2393
+ "epoch": 5.26,
2394
+ "learning_rate": 1.7130338983050848e-05,
2395
+ "loss": 0.0006,
2396
+ "step": 194000
2397
+ },
2398
+ {
2399
+ "epoch": 5.27,
2400
+ "learning_rate": 1.7045593220338985e-05,
2401
+ "loss": 0.0006,
2402
+ "step": 194500
2403
+ },
2404
+ {
2405
+ "epoch": 5.29,
2406
+ "learning_rate": 1.6960847457627118e-05,
2407
+ "loss": 0.0006,
2408
+ "step": 195000
2409
+ },
2410
+ {
2411
+ "epoch": 5.3,
2412
+ "learning_rate": 1.687627118644068e-05,
2413
+ "loss": 0.0005,
2414
+ "step": 195500
2415
+ },
2416
+ {
2417
+ "epoch": 5.32,
2418
+ "learning_rate": 1.6791525423728816e-05,
2419
+ "loss": 0.0006,
2420
+ "step": 196000
2421
+ },
2422
+ {
2423
+ "epoch": 5.33,
2424
+ "learning_rate": 1.670677966101695e-05,
2425
+ "loss": 0.0006,
2426
+ "step": 196500
2427
+ },
2428
+ {
2429
+ "epoch": 5.34,
2430
+ "learning_rate": 1.6622033898305086e-05,
2431
+ "loss": 0.0006,
2432
+ "step": 197000
2433
+ },
2434
+ {
2435
+ "epoch": 5.36,
2436
+ "learning_rate": 1.653728813559322e-05,
2437
+ "loss": 0.0006,
2438
+ "step": 197500
2439
+ },
2440
+ {
2441
+ "epoch": 5.37,
2442
+ "learning_rate": 1.645254237288136e-05,
2443
+ "loss": 0.0005,
2444
+ "step": 198000
2445
+ },
2446
+ {
2447
+ "epoch": 5.38,
2448
+ "learning_rate": 1.6367796610169493e-05,
2449
+ "loss": 0.0007,
2450
+ "step": 198500
2451
+ },
2452
+ {
2453
+ "epoch": 5.4,
2454
+ "learning_rate": 1.6283389830508473e-05,
2455
+ "loss": 0.0007,
2456
+ "step": 199000
2457
+ },
2458
+ {
2459
+ "epoch": 5.41,
2460
+ "learning_rate": 1.619864406779661e-05,
2461
+ "loss": 0.0006,
2462
+ "step": 199500
2463
+ },
2464
+ {
2465
+ "epoch": 5.42,
2466
+ "learning_rate": 1.6113898305084747e-05,
2467
+ "loss": 0.0006,
2468
+ "step": 200000
2469
+ },
2470
+ {
2471
+ "epoch": 5.44,
2472
+ "learning_rate": 1.6029152542372883e-05,
2473
+ "loss": 0.0005,
2474
+ "step": 200500
2475
+ },
2476
+ {
2477
+ "epoch": 5.45,
2478
+ "learning_rate": 1.5944406779661016e-05,
2479
+ "loss": 0.0005,
2480
+ "step": 201000
2481
+ },
2482
+ {
2483
+ "epoch": 5.46,
2484
+ "learning_rate": 1.5859661016949153e-05,
2485
+ "loss": 0.0006,
2486
+ "step": 201500
2487
+ },
2488
+ {
2489
+ "epoch": 5.48,
2490
+ "learning_rate": 1.5775084745762715e-05,
2491
+ "loss": 0.0006,
2492
+ "step": 202000
2493
+ },
2494
+ {
2495
+ "epoch": 5.49,
2496
+ "learning_rate": 1.5690338983050848e-05,
2497
+ "loss": 0.0004,
2498
+ "step": 202500
2499
+ },
2500
+ {
2501
+ "epoch": 5.51,
2502
+ "learning_rate": 1.5605593220338985e-05,
2503
+ "loss": 0.0005,
2504
+ "step": 203000
2505
+ },
2506
+ {
2507
+ "epoch": 5.52,
2508
+ "learning_rate": 1.5520847457627118e-05,
2509
+ "loss": 0.0006,
2510
+ "step": 203500
2511
+ },
2512
+ {
2513
+ "epoch": 5.53,
2514
+ "learning_rate": 1.5436101694915255e-05,
2515
+ "loss": 0.0005,
2516
+ "step": 204000
2517
+ },
2518
+ {
2519
+ "epoch": 5.55,
2520
+ "learning_rate": 1.535135593220339e-05,
2521
+ "loss": 0.0006,
2522
+ "step": 204500
2523
+ },
2524
+ {
2525
+ "epoch": 5.56,
2526
+ "learning_rate": 1.5266610169491528e-05,
2527
+ "loss": 0.0005,
2528
+ "step": 205000
2529
+ },
2530
+ {
2531
+ "epoch": 5.57,
2532
+ "learning_rate": 1.5181864406779661e-05,
2533
+ "loss": 0.0005,
2534
+ "step": 205500
2535
+ },
2536
+ {
2537
+ "epoch": 5.59,
2538
+ "learning_rate": 1.5097118644067797e-05,
2539
+ "loss": 0.0004,
2540
+ "step": 206000
2541
+ },
2542
+ {
2543
+ "epoch": 5.6,
2544
+ "learning_rate": 1.5012372881355932e-05,
2545
+ "loss": 0.0005,
2546
+ "step": 206500
2547
+ },
2548
+ {
2549
+ "epoch": 5.61,
2550
+ "learning_rate": 1.4927627118644069e-05,
2551
+ "loss": 0.0008,
2552
+ "step": 207000
2553
+ },
2554
+ {
2555
+ "epoch": 5.63,
2556
+ "learning_rate": 1.4842881355932204e-05,
2557
+ "loss": 0.0006,
2558
+ "step": 207500
2559
+ },
2560
+ {
2561
+ "epoch": 5.64,
2562
+ "learning_rate": 1.475813559322034e-05,
2563
+ "loss": 0.0004,
2564
+ "step": 208000
2565
+ },
2566
+ {
2567
+ "epoch": 5.65,
2568
+ "learning_rate": 1.46735593220339e-05,
2569
+ "loss": 0.0005,
2570
+ "step": 208500
2571
+ },
2572
+ {
2573
+ "epoch": 5.67,
2574
+ "learning_rate": 1.4588813559322034e-05,
2575
+ "loss": 0.0006,
2576
+ "step": 209000
2577
+ },
2578
+ {
2579
+ "epoch": 5.68,
2580
+ "learning_rate": 1.450406779661017e-05,
2581
+ "loss": 0.0006,
2582
+ "step": 209500
2583
+ },
2584
+ {
2585
+ "epoch": 5.69,
2586
+ "learning_rate": 1.4419322033898305e-05,
2587
+ "loss": 0.0006,
2588
+ "step": 210000
2589
+ },
2590
+ {
2591
+ "epoch": 5.71,
2592
+ "learning_rate": 1.4334576271186442e-05,
2593
+ "loss": 0.0005,
2594
+ "step": 210500
2595
+ },
2596
+ {
2597
+ "epoch": 5.72,
2598
+ "learning_rate": 1.4249999999999999e-05,
2599
+ "loss": 0.0004,
2600
+ "step": 211000
2601
+ },
2602
+ {
2603
+ "epoch": 5.74,
2604
+ "learning_rate": 1.4165254237288136e-05,
2605
+ "loss": 0.0004,
2606
+ "step": 211500
2607
+ },
2608
+ {
2609
+ "epoch": 5.75,
2610
+ "learning_rate": 1.4080677966101696e-05,
2611
+ "loss": 0.0007,
2612
+ "step": 212000
2613
+ },
2614
+ {
2615
+ "epoch": 5.76,
2616
+ "learning_rate": 1.3995932203389831e-05,
2617
+ "loss": 0.0004,
2618
+ "step": 212500
2619
+ },
2620
+ {
2621
+ "epoch": 5.78,
2622
+ "learning_rate": 1.3911186440677968e-05,
2623
+ "loss": 0.0005,
2624
+ "step": 213000
2625
+ },
2626
+ {
2627
+ "epoch": 5.79,
2628
+ "learning_rate": 1.38264406779661e-05,
2629
+ "loss": 0.0005,
2630
+ "step": 213500
2631
+ },
2632
+ {
2633
+ "epoch": 5.8,
2634
+ "learning_rate": 1.3741694915254239e-05,
2635
+ "loss": 0.0005,
2636
+ "step": 214000
2637
+ },
2638
+ {
2639
+ "epoch": 5.82,
2640
+ "learning_rate": 1.3656949152542372e-05,
2641
+ "loss": 0.0005,
2642
+ "step": 214500
2643
+ },
2644
+ {
2645
+ "epoch": 5.83,
2646
+ "learning_rate": 1.3572372881355933e-05,
2647
+ "loss": 0.0006,
2648
+ "step": 215000
2649
+ },
2650
+ {
2651
+ "epoch": 5.84,
2652
+ "learning_rate": 1.348762711864407e-05,
2653
+ "loss": 0.0006,
2654
+ "step": 215500
2655
+ },
2656
+ {
2657
+ "epoch": 5.86,
2658
+ "learning_rate": 1.3402881355932204e-05,
2659
+ "loss": 0.0005,
2660
+ "step": 216000
2661
+ },
2662
+ {
2663
+ "epoch": 5.87,
2664
+ "learning_rate": 1.331813559322034e-05,
2665
+ "loss": 0.0006,
2666
+ "step": 216500
2667
+ },
2668
+ {
2669
+ "epoch": 5.88,
2670
+ "learning_rate": 1.3233389830508476e-05,
2671
+ "loss": 0.0004,
2672
+ "step": 217000
2673
+ },
2674
+ {
2675
+ "epoch": 5.9,
2676
+ "learning_rate": 1.3148644067796612e-05,
2677
+ "loss": 0.0006,
2678
+ "step": 217500
2679
+ },
2680
+ {
2681
+ "epoch": 5.91,
2682
+ "learning_rate": 1.3063898305084745e-05,
2683
+ "loss": 0.0005,
2684
+ "step": 218000
2685
+ },
2686
+ {
2687
+ "epoch": 5.93,
2688
+ "learning_rate": 1.2979152542372882e-05,
2689
+ "loss": 0.0005,
2690
+ "step": 218500
2691
+ },
2692
+ {
2693
+ "epoch": 5.94,
2694
+ "learning_rate": 1.2894406779661017e-05,
2695
+ "loss": 0.0006,
2696
+ "step": 219000
2697
+ },
2698
+ {
2699
+ "epoch": 5.95,
2700
+ "learning_rate": 1.2809661016949153e-05,
2701
+ "loss": 0.0005,
2702
+ "step": 219500
2703
+ },
2704
+ {
2705
+ "epoch": 5.97,
2706
+ "learning_rate": 1.2724915254237288e-05,
2707
+ "loss": 0.0006,
2708
+ "step": 220000
2709
+ },
2710
+ {
2711
+ "epoch": 5.98,
2712
+ "learning_rate": 1.2640338983050849e-05,
2713
+ "loss": 0.0004,
2714
+ "step": 220500
2715
+ },
2716
+ {
2717
+ "epoch": 5.99,
2718
+ "learning_rate": 1.2555762711864407e-05,
2719
+ "loss": 0.0006,
2720
+ "step": 221000
2721
+ },
2722
+ {
2723
+ "epoch": 6.0,
2724
+ "eval_accuracy": 0.9993843762384619,
2725
+ "eval_f1": 0.0,
2726
+ "eval_loss": 0.0035067975986748934,
2727
+ "eval_precision": 0.0,
2728
+ "eval_recall": 0.0,
2729
+ "eval_runtime": 363.9269,
2730
+ "eval_samples_per_second": 108.838,
2731
+ "eval_steps_per_second": 6.804,
2732
+ "step": 221250
2733
+ },
2734
+ {
2735
+ "epoch": 6.01,
2736
+ "learning_rate": 1.2471016949152544e-05,
2737
+ "loss": 0.0003,
2738
+ "step": 221500
2739
+ },
2740
+ {
2741
+ "epoch": 6.02,
2742
+ "learning_rate": 1.2386271186440679e-05,
2743
+ "loss": 0.0003,
2744
+ "step": 222000
2745
+ },
2746
+ {
2747
+ "epoch": 6.03,
2748
+ "learning_rate": 1.2301525423728814e-05,
2749
+ "loss": 0.0004,
2750
+ "step": 222500
2751
+ },
2752
+ {
2753
+ "epoch": 6.05,
2754
+ "learning_rate": 1.221677966101695e-05,
2755
+ "loss": 0.0003,
2756
+ "step": 223000
2757
+ },
2758
+ {
2759
+ "epoch": 6.06,
2760
+ "learning_rate": 1.2132033898305085e-05,
2761
+ "loss": 0.0003,
2762
+ "step": 223500
2763
+ },
2764
+ {
2765
+ "epoch": 6.07,
2766
+ "learning_rate": 1.2047457627118644e-05,
2767
+ "loss": 0.0004,
2768
+ "step": 224000
2769
+ },
2770
+ {
2771
+ "epoch": 6.09,
2772
+ "learning_rate": 1.1962711864406779e-05,
2773
+ "loss": 0.0004,
2774
+ "step": 224500
2775
+ },
2776
+ {
2777
+ "epoch": 6.1,
2778
+ "learning_rate": 1.1877966101694917e-05,
2779
+ "loss": 0.0004,
2780
+ "step": 225000
2781
+ },
2782
+ {
2783
+ "epoch": 6.12,
2784
+ "learning_rate": 1.1793220338983052e-05,
2785
+ "loss": 0.0004,
2786
+ "step": 225500
2787
+ },
2788
+ {
2789
+ "epoch": 6.13,
2790
+ "learning_rate": 1.170864406779661e-05,
2791
+ "loss": 0.0004,
2792
+ "step": 226000
2793
+ },
2794
+ {
2795
+ "epoch": 6.14,
2796
+ "learning_rate": 1.1623898305084746e-05,
2797
+ "loss": 0.0004,
2798
+ "step": 226500
2799
+ },
2800
+ {
2801
+ "epoch": 6.16,
2802
+ "learning_rate": 1.1539152542372882e-05,
2803
+ "loss": 0.0004,
2804
+ "step": 227000
2805
+ },
2806
+ {
2807
+ "epoch": 6.17,
2808
+ "learning_rate": 1.1454406779661017e-05,
2809
+ "loss": 0.0004,
2810
+ "step": 227500
2811
+ },
2812
+ {
2813
+ "epoch": 6.18,
2814
+ "learning_rate": 1.1369661016949152e-05,
2815
+ "loss": 0.0004,
2816
+ "step": 228000
2817
+ },
2818
+ {
2819
+ "epoch": 6.2,
2820
+ "learning_rate": 1.1284915254237288e-05,
2821
+ "loss": 0.0004,
2822
+ "step": 228500
2823
+ },
2824
+ {
2825
+ "epoch": 6.21,
2826
+ "learning_rate": 1.1200338983050849e-05,
2827
+ "loss": 0.0003,
2828
+ "step": 229000
2829
+ },
2830
+ {
2831
+ "epoch": 6.22,
2832
+ "learning_rate": 1.1115593220338984e-05,
2833
+ "loss": 0.0004,
2834
+ "step": 229500
2835
+ },
2836
+ {
2837
+ "epoch": 6.24,
2838
+ "learning_rate": 1.1030847457627119e-05,
2839
+ "loss": 0.0003,
2840
+ "step": 230000
2841
+ },
2842
+ {
2843
+ "epoch": 6.25,
2844
+ "learning_rate": 1.0946101694915255e-05,
2845
+ "loss": 0.0004,
2846
+ "step": 230500
2847
+ },
2848
+ {
2849
+ "epoch": 6.26,
2850
+ "learning_rate": 1.086135593220339e-05,
2851
+ "loss": 0.0004,
2852
+ "step": 231000
2853
+ },
2854
+ {
2855
+ "epoch": 6.28,
2856
+ "learning_rate": 1.0776610169491527e-05,
2857
+ "loss": 0.0005,
2858
+ "step": 231500
2859
+ },
2860
+ {
2861
+ "epoch": 6.29,
2862
+ "learning_rate": 1.0691864406779662e-05,
2863
+ "loss": 0.0002,
2864
+ "step": 232000
2865
+ },
2866
+ {
2867
+ "epoch": 6.31,
2868
+ "learning_rate": 1.0607118644067796e-05,
2869
+ "loss": 0.0004,
2870
+ "step": 232500
2871
+ },
2872
+ {
2873
+ "epoch": 6.32,
2874
+ "learning_rate": 1.0522372881355933e-05,
2875
+ "loss": 0.0004,
2876
+ "step": 233000
2877
+ },
2878
+ {
2879
+ "epoch": 6.33,
2880
+ "learning_rate": 1.0437627118644068e-05,
2881
+ "loss": 0.0003,
2882
+ "step": 233500
2883
+ },
2884
+ {
2885
+ "epoch": 6.35,
2886
+ "learning_rate": 1.0352881355932203e-05,
2887
+ "loss": 0.0004,
2888
+ "step": 234000
2889
+ },
2890
+ {
2891
+ "epoch": 6.36,
2892
+ "learning_rate": 1.0268305084745763e-05,
2893
+ "loss": 0.0004,
2894
+ "step": 234500
2895
+ },
2896
+ {
2897
+ "epoch": 6.37,
2898
+ "learning_rate": 1.01835593220339e-05,
2899
+ "loss": 0.0003,
2900
+ "step": 235000
2901
+ },
2902
+ {
2903
+ "epoch": 6.39,
2904
+ "learning_rate": 1.0098813559322035e-05,
2905
+ "loss": 0.0006,
2906
+ "step": 235500
2907
+ },
2908
+ {
2909
+ "epoch": 6.4,
2910
+ "learning_rate": 1.001406779661017e-05,
2911
+ "loss": 0.0003,
2912
+ "step": 236000
2913
+ },
2914
+ {
2915
+ "epoch": 6.41,
2916
+ "learning_rate": 9.929322033898306e-06,
2917
+ "loss": 0.0004,
2918
+ "step": 236500
2919
+ },
2920
+ {
2921
+ "epoch": 6.43,
2922
+ "learning_rate": 9.844576271186441e-06,
2923
+ "loss": 0.0004,
2924
+ "step": 237000
2925
+ },
2926
+ {
2927
+ "epoch": 6.44,
2928
+ "learning_rate": 9.760000000000001e-06,
2929
+ "loss": 0.0003,
2930
+ "step": 237500
2931
+ },
2932
+ {
2933
+ "epoch": 6.45,
2934
+ "learning_rate": 9.675254237288136e-06,
2935
+ "loss": 0.0004,
2936
+ "step": 238000
2937
+ },
2938
+ {
2939
+ "epoch": 6.47,
2940
+ "learning_rate": 9.590508474576273e-06,
2941
+ "loss": 0.0003,
2942
+ "step": 238500
2943
+ },
2944
+ {
2945
+ "epoch": 6.48,
2946
+ "learning_rate": 9.505762711864408e-06,
2947
+ "loss": 0.0003,
2948
+ "step": 239000
2949
+ },
2950
+ {
2951
+ "epoch": 6.49,
2952
+ "learning_rate": 9.421016949152543e-06,
2953
+ "loss": 0.0003,
2954
+ "step": 239500
2955
+ },
2956
+ {
2957
+ "epoch": 6.51,
2958
+ "learning_rate": 9.33627118644068e-06,
2959
+ "loss": 0.0003,
2960
+ "step": 240000
2961
+ },
2962
+ {
2963
+ "epoch": 6.52,
2964
+ "learning_rate": 9.251525423728814e-06,
2965
+ "loss": 0.0003,
2966
+ "step": 240500
2967
+ },
2968
+ {
2969
+ "epoch": 6.54,
2970
+ "learning_rate": 9.16677966101695e-06,
2971
+ "loss": 0.0004,
2972
+ "step": 241000
2973
+ },
2974
+ {
2975
+ "epoch": 6.55,
2976
+ "learning_rate": 9.082033898305086e-06,
2977
+ "loss": 0.0004,
2978
+ "step": 241500
2979
+ },
2980
+ {
2981
+ "epoch": 6.56,
2982
+ "learning_rate": 8.997457627118644e-06,
2983
+ "loss": 0.0005,
2984
+ "step": 242000
2985
+ },
2986
+ {
2987
+ "epoch": 6.58,
2988
+ "learning_rate": 8.912881355932203e-06,
2989
+ "loss": 0.0004,
2990
+ "step": 242500
2991
+ },
2992
+ {
2993
+ "epoch": 6.59,
2994
+ "learning_rate": 8.82813559322034e-06,
2995
+ "loss": 0.0003,
2996
+ "step": 243000
2997
+ },
2998
+ {
2999
+ "epoch": 6.6,
3000
+ "learning_rate": 8.743389830508474e-06,
3001
+ "loss": 0.0006,
3002
+ "step": 243500
3003
+ },
3004
+ {
3005
+ "epoch": 6.62,
3006
+ "learning_rate": 8.658644067796611e-06,
3007
+ "loss": 0.0004,
3008
+ "step": 244000
3009
+ },
3010
+ {
3011
+ "epoch": 6.63,
3012
+ "learning_rate": 8.57406779661017e-06,
3013
+ "loss": 0.0003,
3014
+ "step": 244500
3015
+ },
3016
+ {
3017
+ "epoch": 6.64,
3018
+ "learning_rate": 8.489322033898306e-06,
3019
+ "loss": 0.0004,
3020
+ "step": 245000
3021
+ },
3022
+ {
3023
+ "epoch": 6.66,
3024
+ "learning_rate": 8.404576271186441e-06,
3025
+ "loss": 0.0002,
3026
+ "step": 245500
3027
+ },
3028
+ {
3029
+ "epoch": 6.67,
3030
+ "learning_rate": 8.319830508474578e-06,
3031
+ "loss": 0.0004,
3032
+ "step": 246000
3033
+ },
3034
+ {
3035
+ "epoch": 6.68,
3036
+ "learning_rate": 8.235084745762713e-06,
3037
+ "loss": 0.0003,
3038
+ "step": 246500
3039
+ },
3040
+ {
3041
+ "epoch": 6.7,
3042
+ "learning_rate": 8.150508474576271e-06,
3043
+ "loss": 0.0003,
3044
+ "step": 247000
3045
+ },
3046
+ {
3047
+ "epoch": 6.71,
3048
+ "learning_rate": 8.065762711864406e-06,
3049
+ "loss": 0.0004,
3050
+ "step": 247500
3051
+ },
3052
+ {
3053
+ "epoch": 6.73,
3054
+ "learning_rate": 7.981016949152543e-06,
3055
+ "loss": 0.0002,
3056
+ "step": 248000
3057
+ },
3058
+ {
3059
+ "epoch": 6.74,
3060
+ "learning_rate": 7.896271186440678e-06,
3061
+ "loss": 0.0003,
3062
+ "step": 248500
3063
+ },
3064
+ {
3065
+ "epoch": 6.75,
3066
+ "learning_rate": 7.811694915254238e-06,
3067
+ "loss": 0.0004,
3068
+ "step": 249000
3069
+ },
3070
+ {
3071
+ "epoch": 6.77,
3072
+ "learning_rate": 7.726949152542373e-06,
3073
+ "loss": 0.0003,
3074
+ "step": 249500
3075
+ },
3076
+ {
3077
+ "epoch": 6.78,
3078
+ "learning_rate": 7.642203389830508e-06,
3079
+ "loss": 0.0003,
3080
+ "step": 250000
3081
+ },
3082
+ {
3083
+ "epoch": 6.79,
3084
+ "learning_rate": 7.5574576271186445e-06,
3085
+ "loss": 0.0003,
3086
+ "step": 250500
3087
+ },
3088
+ {
3089
+ "epoch": 6.81,
3090
+ "learning_rate": 7.472711864406779e-06,
3091
+ "loss": 0.0004,
3092
+ "step": 251000
3093
+ },
3094
+ {
3095
+ "epoch": 6.82,
3096
+ "learning_rate": 7.387966101694915e-06,
3097
+ "loss": 0.0003,
3098
+ "step": 251500
3099
+ },
3100
+ {
3101
+ "epoch": 6.83,
3102
+ "learning_rate": 7.303220338983051e-06,
3103
+ "loss": 0.0003,
3104
+ "step": 252000
3105
+ },
3106
+ {
3107
+ "epoch": 6.85,
3108
+ "learning_rate": 7.218474576271187e-06,
3109
+ "loss": 0.0003,
3110
+ "step": 252500
3111
+ },
3112
+ {
3113
+ "epoch": 6.86,
3114
+ "learning_rate": 7.133728813559322e-06,
3115
+ "loss": 0.0004,
3116
+ "step": 253000
3117
+ },
3118
+ {
3119
+ "epoch": 6.87,
3120
+ "learning_rate": 7.048983050847457e-06,
3121
+ "loss": 0.0002,
3122
+ "step": 253500
3123
+ },
3124
+ {
3125
+ "epoch": 6.89,
3126
+ "learning_rate": 6.964237288135593e-06,
3127
+ "loss": 0.0002,
3128
+ "step": 254000
3129
+ },
3130
+ {
3131
+ "epoch": 6.9,
3132
+ "learning_rate": 6.879491525423729e-06,
3133
+ "loss": 0.0003,
3134
+ "step": 254500
3135
+ },
3136
+ {
3137
+ "epoch": 6.92,
3138
+ "learning_rate": 6.794745762711864e-06,
3139
+ "loss": 0.0003,
3140
+ "step": 255000
3141
+ },
3142
+ {
3143
+ "epoch": 6.93,
3144
+ "learning_rate": 6.710338983050849e-06,
3145
+ "loss": 0.0003,
3146
+ "step": 255500
3147
+ },
3148
+ {
3149
+ "epoch": 6.94,
3150
+ "learning_rate": 6.6255932203389835e-06,
3151
+ "loss": 0.0002,
3152
+ "step": 256000
3153
+ },
3154
+ {
3155
+ "epoch": 6.96,
3156
+ "learning_rate": 6.540847457627119e-06,
3157
+ "loss": 0.0003,
3158
+ "step": 256500
3159
+ },
3160
+ {
3161
+ "epoch": 6.97,
3162
+ "learning_rate": 6.456101694915255e-06,
3163
+ "loss": 0.0004,
3164
+ "step": 257000
3165
+ },
3166
+ {
3167
+ "epoch": 6.98,
3168
+ "learning_rate": 6.371355932203391e-06,
3169
+ "loss": 0.0003,
3170
+ "step": 257500
3171
+ },
3172
+ {
3173
+ "epoch": 7.0,
3174
+ "learning_rate": 6.2866101694915265e-06,
3175
+ "loss": 0.0004,
3176
+ "step": 258000
3177
+ },
3178
+ {
3179
+ "epoch": 7.0,
3180
+ "eval_accuracy": 0.9993948198558451,
3181
+ "eval_f1": 0.0,
3182
+ "eval_loss": 0.003731178119778633,
3183
+ "eval_precision": 0.0,
3184
+ "eval_recall": 0.0,
3185
+ "eval_runtime": 371.4536,
3186
+ "eval_samples_per_second": 106.632,
3187
+ "eval_steps_per_second": 6.666,
3188
+ "step": 258125
3189
+ },
3190
+ {
3191
+ "epoch": 7.01,
3192
+ "learning_rate": 6.201864406779661e-06,
3193
+ "loss": 0.0003,
3194
+ "step": 258500
3195
+ },
3196
+ {
3197
+ "epoch": 7.02,
3198
+ "learning_rate": 6.117118644067797e-06,
3199
+ "loss": 0.0002,
3200
+ "step": 259000
3201
+ },
3202
+ {
3203
+ "epoch": 7.04,
3204
+ "learning_rate": 6.032372881355933e-06,
3205
+ "loss": 0.0002,
3206
+ "step": 259500
3207
+ },
3208
+ {
3209
+ "epoch": 7.05,
3210
+ "learning_rate": 5.9477966101694915e-06,
3211
+ "loss": 0.0002,
3212
+ "step": 260000
3213
+ },
3214
+ {
3215
+ "epoch": 7.06,
3216
+ "learning_rate": 5.863050847457627e-06,
3217
+ "loss": 0.0003,
3218
+ "step": 260500
3219
+ },
3220
+ {
3221
+ "epoch": 7.08,
3222
+ "learning_rate": 5.778305084745763e-06,
3223
+ "loss": 0.0002,
3224
+ "step": 261000
3225
+ },
3226
+ {
3227
+ "epoch": 7.09,
3228
+ "learning_rate": 5.6937288135593226e-06,
3229
+ "loss": 0.0003,
3230
+ "step": 261500
3231
+ },
3232
+ {
3233
+ "epoch": 7.11,
3234
+ "learning_rate": 5.6089830508474574e-06,
3235
+ "loss": 0.0002,
3236
+ "step": 262000
3237
+ },
3238
+ {
3239
+ "epoch": 7.12,
3240
+ "learning_rate": 5.524237288135593e-06,
3241
+ "loss": 0.0002,
3242
+ "step": 262500
3243
+ },
3244
+ {
3245
+ "epoch": 7.13,
3246
+ "learning_rate": 5.439661016949153e-06,
3247
+ "loss": 0.0002,
3248
+ "step": 263000
3249
+ },
3250
+ {
3251
+ "epoch": 7.15,
3252
+ "learning_rate": 5.3549152542372885e-06,
3253
+ "loss": 0.0002,
3254
+ "step": 263500
3255
+ },
3256
+ {
3257
+ "epoch": 7.16,
3258
+ "learning_rate": 5.270169491525423e-06,
3259
+ "loss": 0.0003,
3260
+ "step": 264000
3261
+ },
3262
+ {
3263
+ "epoch": 7.17,
3264
+ "learning_rate": 5.18542372881356e-06,
3265
+ "loss": 0.0001,
3266
+ "step": 264500
3267
+ },
3268
+ {
3269
+ "epoch": 7.19,
3270
+ "learning_rate": 5.100677966101696e-06,
3271
+ "loss": 0.0003,
3272
+ "step": 265000
3273
+ },
3274
+ {
3275
+ "epoch": 7.2,
3276
+ "learning_rate": 5.0159322033898305e-06,
3277
+ "loss": 0.0003,
3278
+ "step": 265500
3279
+ },
3280
+ {
3281
+ "epoch": 7.21,
3282
+ "learning_rate": 4.931186440677966e-06,
3283
+ "loss": 0.0002,
3284
+ "step": 266000
3285
+ },
3286
+ {
3287
+ "epoch": 7.23,
3288
+ "learning_rate": 4.846440677966102e-06,
3289
+ "loss": 0.0002,
3290
+ "step": 266500
3291
+ },
3292
+ {
3293
+ "epoch": 7.24,
3294
+ "learning_rate": 4.761694915254238e-06,
3295
+ "loss": 0.0002,
3296
+ "step": 267000
3297
+ },
3298
+ {
3299
+ "epoch": 7.25,
3300
+ "learning_rate": 4.6769491525423735e-06,
3301
+ "loss": 0.0002,
3302
+ "step": 267500
3303
+ },
3304
+ {
3305
+ "epoch": 7.27,
3306
+ "learning_rate": 4.592372881355932e-06,
3307
+ "loss": 0.0004,
3308
+ "step": 268000
3309
+ },
3310
+ {
3311
+ "epoch": 7.28,
3312
+ "learning_rate": 4.507627118644068e-06,
3313
+ "loss": 0.0002,
3314
+ "step": 268500
3315
+ },
3316
+ {
3317
+ "epoch": 7.29,
3318
+ "learning_rate": 4.422881355932204e-06,
3319
+ "loss": 0.0002,
3320
+ "step": 269000
3321
+ },
3322
+ {
3323
+ "epoch": 7.31,
3324
+ "learning_rate": 4.3381355932203385e-06,
3325
+ "loss": 0.0002,
3326
+ "step": 269500
3327
+ },
3328
+ {
3329
+ "epoch": 7.32,
3330
+ "learning_rate": 4.253389830508474e-06,
3331
+ "loss": 0.0002,
3332
+ "step": 270000
3333
+ },
3334
+ {
3335
+ "epoch": 7.34,
3336
+ "learning_rate": 4.168644067796611e-06,
3337
+ "loss": 0.0002,
3338
+ "step": 270500
3339
+ },
3340
+ {
3341
+ "epoch": 7.35,
3342
+ "learning_rate": 4.0840677966101696e-06,
3343
+ "loss": 0.0003,
3344
+ "step": 271000
3345
+ },
3346
+ {
3347
+ "epoch": 7.36,
3348
+ "learning_rate": 3.999322033898305e-06,
3349
+ "loss": 0.0003,
3350
+ "step": 271500
3351
+ },
3352
+ {
3353
+ "epoch": 7.38,
3354
+ "learning_rate": 3.914576271186441e-06,
3355
+ "loss": 0.0002,
3356
+ "step": 272000
3357
+ },
3358
+ {
3359
+ "epoch": 7.39,
3360
+ "learning_rate": 3.829830508474577e-06,
3361
+ "loss": 0.0002,
3362
+ "step": 272500
3363
+ },
3364
+ {
3365
+ "epoch": 7.4,
3366
+ "learning_rate": 3.745084745762712e-06,
3367
+ "loss": 0.0002,
3368
+ "step": 273000
3369
+ },
3370
+ {
3371
+ "epoch": 7.42,
3372
+ "learning_rate": 3.660338983050848e-06,
3373
+ "loss": 0.0002,
3374
+ "step": 273500
3375
+ },
3376
+ {
3377
+ "epoch": 7.43,
3378
+ "learning_rate": 3.575593220338983e-06,
3379
+ "loss": 0.0003,
3380
+ "step": 274000
3381
+ },
3382
+ {
3383
+ "epoch": 7.44,
3384
+ "learning_rate": 3.490847457627119e-06,
3385
+ "loss": 0.0003,
3386
+ "step": 274500
3387
+ },
3388
+ {
3389
+ "epoch": 7.46,
3390
+ "learning_rate": 3.406101694915254e-06,
3391
+ "loss": 0.0003,
3392
+ "step": 275000
3393
+ },
3394
+ {
3395
+ "epoch": 7.47,
3396
+ "learning_rate": 3.3215254237288133e-06,
3397
+ "loss": 0.0002,
3398
+ "step": 275500
3399
+ },
3400
+ {
3401
+ "epoch": 7.48,
3402
+ "learning_rate": 3.236779661016949e-06,
3403
+ "loss": 0.0002,
3404
+ "step": 276000
3405
+ },
3406
+ {
3407
+ "epoch": 7.5,
3408
+ "learning_rate": 3.1520338983050847e-06,
3409
+ "loss": 0.0002,
3410
+ "step": 276500
3411
+ },
3412
+ {
3413
+ "epoch": 7.51,
3414
+ "learning_rate": 3.0672881355932205e-06,
3415
+ "loss": 0.0002,
3416
+ "step": 277000
3417
+ },
3418
+ {
3419
+ "epoch": 7.53,
3420
+ "learning_rate": 2.9827118644067796e-06,
3421
+ "loss": 0.0002,
3422
+ "step": 277500
3423
+ },
3424
+ {
3425
+ "epoch": 7.54,
3426
+ "learning_rate": 2.8979661016949154e-06,
3427
+ "loss": 0.0001,
3428
+ "step": 278000
3429
+ },
3430
+ {
3431
+ "epoch": 7.55,
3432
+ "learning_rate": 2.813220338983051e-06,
3433
+ "loss": 0.0003,
3434
+ "step": 278500
3435
+ },
3436
+ {
3437
+ "epoch": 7.57,
3438
+ "learning_rate": 2.728474576271187e-06,
3439
+ "loss": 0.0002,
3440
+ "step": 279000
3441
+ },
3442
+ {
3443
+ "epoch": 7.58,
3444
+ "learning_rate": 2.643898305084746e-06,
3445
+ "loss": 0.0002,
3446
+ "step": 279500
3447
+ },
3448
+ {
3449
+ "epoch": 7.59,
3450
+ "learning_rate": 2.559322033898305e-06,
3451
+ "loss": 0.0002,
3452
+ "step": 280000
3453
+ },
3454
+ {
3455
+ "epoch": 7.61,
3456
+ "learning_rate": 2.474576271186441e-06,
3457
+ "loss": 0.0002,
3458
+ "step": 280500
3459
+ },
3460
+ {
3461
+ "epoch": 7.62,
3462
+ "learning_rate": 2.3898305084745766e-06,
3463
+ "loss": 0.0002,
3464
+ "step": 281000
3465
+ },
3466
+ {
3467
+ "epoch": 7.63,
3468
+ "learning_rate": 2.305084745762712e-06,
3469
+ "loss": 0.0002,
3470
+ "step": 281500
3471
+ },
3472
+ {
3473
+ "epoch": 7.65,
3474
+ "learning_rate": 2.2203389830508476e-06,
3475
+ "loss": 0.0002,
3476
+ "step": 282000
3477
+ },
3478
+ {
3479
+ "epoch": 7.66,
3480
+ "learning_rate": 2.1355932203389833e-06,
3481
+ "loss": 0.0002,
3482
+ "step": 282500
3483
+ },
3484
+ {
3485
+ "epoch": 7.67,
3486
+ "learning_rate": 2.0508474576271186e-06,
3487
+ "loss": 0.0001,
3488
+ "step": 283000
3489
+ },
3490
+ {
3491
+ "epoch": 7.69,
3492
+ "learning_rate": 1.9661016949152544e-06,
3493
+ "loss": 0.0002,
3494
+ "step": 283500
3495
+ },
3496
+ {
3497
+ "epoch": 7.7,
3498
+ "learning_rate": 1.8813559322033897e-06,
3499
+ "loss": 0.0002,
3500
+ "step": 284000
3501
+ },
3502
+ {
3503
+ "epoch": 7.72,
3504
+ "learning_rate": 1.7967796610169492e-06,
3505
+ "loss": 0.0002,
3506
+ "step": 284500
3507
+ },
3508
+ {
3509
+ "epoch": 7.73,
3510
+ "learning_rate": 1.7120338983050848e-06,
3511
+ "loss": 0.0001,
3512
+ "step": 285000
3513
+ },
3514
+ {
3515
+ "epoch": 7.74,
3516
+ "learning_rate": 1.6272881355932205e-06,
3517
+ "loss": 0.0002,
3518
+ "step": 285500
3519
+ },
3520
+ {
3521
+ "epoch": 7.76,
3522
+ "learning_rate": 1.542542372881356e-06,
3523
+ "loss": 0.0002,
3524
+ "step": 286000
3525
+ },
3526
+ {
3527
+ "epoch": 7.77,
3528
+ "learning_rate": 1.4579661016949154e-06,
3529
+ "loss": 0.0001,
3530
+ "step": 286500
3531
+ },
3532
+ {
3533
+ "epoch": 7.78,
3534
+ "learning_rate": 1.3733898305084745e-06,
3535
+ "loss": 0.0002,
3536
+ "step": 287000
3537
+ },
3538
+ {
3539
+ "epoch": 7.8,
3540
+ "learning_rate": 1.2886440677966103e-06,
3541
+ "loss": 0.0002,
3542
+ "step": 287500
3543
+ },
3544
+ {
3545
+ "epoch": 7.81,
3546
+ "learning_rate": 1.2038983050847458e-06,
3547
+ "loss": 0.0003,
3548
+ "step": 288000
3549
+ },
3550
+ {
3551
+ "epoch": 7.82,
3552
+ "learning_rate": 1.1191525423728813e-06,
3553
+ "loss": 0.0001,
3554
+ "step": 288500
3555
+ },
3556
+ {
3557
+ "epoch": 7.84,
3558
+ "learning_rate": 1.034406779661017e-06,
3559
+ "loss": 0.0002,
3560
+ "step": 289000
3561
+ },
3562
+ {
3563
+ "epoch": 7.85,
3564
+ "learning_rate": 9.496610169491526e-07,
3565
+ "loss": 0.0001,
3566
+ "step": 289500
3567
+ },
3568
+ {
3569
+ "epoch": 7.86,
3570
+ "learning_rate": 8.649152542372882e-07,
3571
+ "loss": 0.0002,
3572
+ "step": 290000
3573
+ },
3574
+ {
3575
+ "epoch": 7.88,
3576
+ "learning_rate": 7.801694915254238e-07,
3577
+ "loss": 0.0002,
3578
+ "step": 290500
3579
+ },
3580
+ {
3581
+ "epoch": 7.89,
3582
+ "learning_rate": 6.954237288135593e-07,
3583
+ "loss": 0.0003,
3584
+ "step": 291000
3585
+ },
3586
+ {
3587
+ "epoch": 7.91,
3588
+ "learning_rate": 6.106779661016949e-07,
3589
+ "loss": 0.0001,
3590
+ "step": 291500
3591
+ },
3592
+ {
3593
+ "epoch": 7.92,
3594
+ "learning_rate": 5.259322033898306e-07,
3595
+ "loss": 0.0002,
3596
+ "step": 292000
3597
+ },
3598
+ {
3599
+ "epoch": 7.93,
3600
+ "learning_rate": 4.411864406779661e-07,
3601
+ "loss": 0.0002,
3602
+ "step": 292500
3603
+ },
3604
+ {
3605
+ "epoch": 7.95,
3606
+ "learning_rate": 3.5661016949152543e-07,
3607
+ "loss": 0.0002,
3608
+ "step": 293000
3609
+ },
3610
+ {
3611
+ "epoch": 7.96,
3612
+ "learning_rate": 2.71864406779661e-07,
3613
+ "loss": 0.0002,
3614
+ "step": 293500
3615
+ },
3616
+ {
3617
+ "epoch": 7.97,
3618
+ "learning_rate": 1.8711864406779662e-07,
3619
+ "loss": 0.0002,
3620
+ "step": 294000
3621
+ },
3622
+ {
3623
+ "epoch": 7.99,
3624
+ "learning_rate": 1.0237288135593222e-07,
3625
+ "loss": 0.0001,
3626
+ "step": 294500
3627
+ },
3628
+ {
3629
+ "epoch": 8.0,
3630
+ "learning_rate": 1.7627118644067797e-08,
3631
+ "loss": 0.0002,
3632
+ "step": 295000
3633
+ },
3634
+ {
3635
+ "epoch": 8.0,
3636
+ "eval_accuracy": 0.9994025151528644,
3637
+ "eval_f1": 0.0,
3638
+ "eval_loss": 0.003991865552961826,
3639
+ "eval_precision": 0.0,
3640
+ "eval_recall": 0.0,
3641
+ "eval_runtime": 383.2603,
3642
+ "eval_samples_per_second": 103.348,
3643
+ "eval_steps_per_second": 6.46,
3644
+ "step": 295000
3645
+ },
3646
+ {
3647
+ "epoch": 8.0,
3648
+ "step": 295000,
3649
+ "total_flos": 1.21405005598464e+18,
3650
+ "train_loss": 0.0015094334459279553,
3651
+ "train_runtime": 87513.7882,
3652
+ "train_samples_per_second": 53.934,
3653
+ "train_steps_per_second": 3.371
3654
+ }
3655
+ ],
3656
+ "max_steps": 295000,
3657
+ "num_train_epochs": 8,
3658
+ "total_flos": 1.21405005598464e+18,
3659
+ "trial_name": null,
3660
+ "trial_params": null
3661
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:087dc14a08f8313a9411212eb74acdf3d8253190cccfc63091d56fc01d4c8123
3
+ size 3055
vocab.txt ADDED
The diff for this file is too large to render. See raw diff