mtapiapacheco commited on
Commit
362ce35
·
verified ·
1 Parent(s): 42ae51b

Delete len2_5120

Browse files

fixing nesting issue

This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. len2_5120/.gitattributes +0 -35
  2. len2_5120/04.pretrain.sh +0 -67
  3. len2_5120/04.pretrain_200k.sh +0 -62
  4. len2_5120/README.md +0 -56
  5. len2_5120/all_results.json +0 -16
  6. len2_5120/checkpoint-191000/config.json +0 -25
  7. len2_5120/checkpoint-191000/generation_config.json +0 -6
  8. len2_5120/checkpoint-191000/model.safetensors +0 -3
  9. len2_5120/checkpoint-191000/optimizer.pt +0 -3
  10. len2_5120/checkpoint-191000/rng_state.pth +0 -3
  11. len2_5120/checkpoint-191000/scheduler.pt +0 -3
  12. len2_5120/checkpoint-191000/trainer_state.json +0 -1165
  13. len2_5120/checkpoint-191000/training_args.bin +0 -3
  14. len2_5120/checkpoint-192000/config.json +0 -25
  15. len2_5120/checkpoint-192000/generation_config.json +0 -6
  16. len2_5120/checkpoint-192000/model.safetensors +0 -3
  17. len2_5120/checkpoint-192000/optimizer.pt +0 -3
  18. len2_5120/checkpoint-192000/rng_state.pth +0 -3
  19. len2_5120/checkpoint-192000/scheduler.pt +0 -3
  20. len2_5120/checkpoint-192000/trainer_state.json +0 -1171
  21. len2_5120/checkpoint-192000/training_args.bin +0 -3
  22. len2_5120/checkpoint-193000/config.json +0 -25
  23. len2_5120/checkpoint-193000/generation_config.json +0 -6
  24. len2_5120/checkpoint-193000/model.safetensors +0 -3
  25. len2_5120/checkpoint-193000/optimizer.pt +0 -3
  26. len2_5120/checkpoint-193000/rng_state.pth +0 -3
  27. len2_5120/checkpoint-193000/scheduler.pt +0 -3
  28. len2_5120/checkpoint-193000/trainer_state.json +0 -1177
  29. len2_5120/checkpoint-193000/training_args.bin +0 -3
  30. len2_5120/checkpoint-194000/config.json +0 -25
  31. len2_5120/checkpoint-194000/generation_config.json +0 -6
  32. len2_5120/checkpoint-194000/model.safetensors +0 -3
  33. len2_5120/checkpoint-194000/optimizer.pt +0 -3
  34. len2_5120/checkpoint-194000/rng_state.pth +0 -3
  35. len2_5120/checkpoint-194000/scheduler.pt +0 -3
  36. len2_5120/checkpoint-194000/trainer_state.json +0 -1183
  37. len2_5120/checkpoint-194000/training_args.bin +0 -3
  38. len2_5120/checkpoint-195000/config.json +0 -25
  39. len2_5120/checkpoint-195000/generation_config.json +0 -6
  40. len2_5120/checkpoint-195000/model.safetensors +0 -3
  41. len2_5120/checkpoint-195000/optimizer.pt +0 -3
  42. len2_5120/checkpoint-195000/rng_state.pth +0 -3
  43. len2_5120/checkpoint-195000/scheduler.pt +0 -3
  44. len2_5120/checkpoint-195000/trainer_state.json +0 -1189
  45. len2_5120/checkpoint-195000/training_args.bin +0 -3
  46. len2_5120/checkpoint-196000/config.json +0 -25
  47. len2_5120/checkpoint-196000/generation_config.json +0 -6
  48. len2_5120/checkpoint-196000/model.safetensors +0 -3
  49. len2_5120/checkpoint-196000/optimizer.pt +0 -3
  50. len2_5120/checkpoint-196000/rng_state.pth +0 -3
len2_5120/.gitattributes DELETED
@@ -1,35 +0,0 @@
1
- *.7z filter=lfs diff=lfs merge=lfs -text
2
- *.arrow filter=lfs diff=lfs merge=lfs -text
3
- *.bin filter=lfs diff=lfs merge=lfs -text
4
- *.bz2 filter=lfs diff=lfs merge=lfs -text
5
- *.ckpt filter=lfs diff=lfs merge=lfs -text
6
- *.ftz filter=lfs diff=lfs merge=lfs -text
7
- *.gz filter=lfs diff=lfs merge=lfs -text
8
- *.h5 filter=lfs diff=lfs merge=lfs -text
9
- *.joblib filter=lfs diff=lfs merge=lfs -text
10
- *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
- *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
- *.model filter=lfs diff=lfs merge=lfs -text
13
- *.msgpack filter=lfs diff=lfs merge=lfs -text
14
- *.npy filter=lfs diff=lfs merge=lfs -text
15
- *.npz filter=lfs diff=lfs merge=lfs -text
16
- *.onnx filter=lfs diff=lfs merge=lfs -text
17
- *.ot filter=lfs diff=lfs merge=lfs -text
18
- *.parquet filter=lfs diff=lfs merge=lfs -text
19
- *.pb filter=lfs diff=lfs merge=lfs -text
20
- *.pickle filter=lfs diff=lfs merge=lfs -text
21
- *.pkl filter=lfs diff=lfs merge=lfs -text
22
- *.pt filter=lfs diff=lfs merge=lfs -text
23
- *.pth filter=lfs diff=lfs merge=lfs -text
24
- *.rar filter=lfs diff=lfs merge=lfs -text
25
- *.safetensors filter=lfs diff=lfs merge=lfs -text
26
- saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
- *.tar.* filter=lfs diff=lfs merge=lfs -text
28
- *.tar filter=lfs diff=lfs merge=lfs -text
29
- *.tflite filter=lfs diff=lfs merge=lfs -text
30
- *.tgz filter=lfs diff=lfs merge=lfs -text
31
- *.wasm filter=lfs diff=lfs merge=lfs -text
32
- *.xz filter=lfs diff=lfs merge=lfs -text
33
- *.zip filter=lfs diff=lfs merge=lfs -text
34
- *.zst filter=lfs diff=lfs merge=lfs -text
35
- *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/04.pretrain.sh DELETED
@@ -1,67 +0,0 @@
1
- #!/usr/bin/env bash
2
- set -e
3
-
4
-
5
- # --- initialize conda ---
6
- source /root/miniconda3/etc/profile.d/conda.sh
7
-
8
- # --- activate env ---
9
- conda activate bpe_v2
10
-
11
- work_dir=/root/NaN/dna-tokenizer
12
- data_dir=${work_dir}/dna-tokenizer-data
13
- cache_dir=${data_dir}/cache
14
-
15
-
16
- # runnum=len2_2048
17
- runnum=len2_5120
18
-
19
- mkdir -p ${work_dir}/pretrain/models/model_${runnum}
20
- output_dir=${work_dir}/pretrain/models/model_${runnum}
21
-
22
-
23
- cp 04.pretrain.sh ${output_dir}
24
- torchrun --nproc_per_node=8 run_mlm.py \
25
- --output_dir ${output_dir} \
26
- --model_type bert \
27
- --tokenizer_name ${work_dir}/merge_bpe/vocab_5120/merge_tokenizer_unigram_len2.json \
28
- --config_name ${work_dir}/BPE_merge/config_5120.json \
29
- --project_name token_eval_len2_5120 \
30
- --do_train True \
31
- --model_max_length 512 \
32
- --max_seq_length 512 \
33
- --line_by_line True \
34
- --pad_to_max_length True \
35
- --train_file ${data_dir}/merge_bpe_5120_allchr_all_tokenized_train_chrOnly.tsv \
36
- --validation_file ${data_dir}/merge_bpe_5120_allchr_all_tokenized_val_chrOnly.tsv \
37
- --cache_dir ${cache_dir} \
38
- --use_fast_tokenizer True \
39
- --do_eval True \
40
- --gradient_accumulation_steps 1 \
41
- --per_device_train_batch_size 96 \
42
- --per_device_eval_batch_size 96 \
43
- --save_steps 1000 \
44
- --save_total_limit 10 \
45
- --max_steps 100000 \
46
- --logging_steps 1000 \
47
- --learning_rate 4e-5 \
48
- --adam_epsilon 1e-6 \
49
- --weight_decay 0.01 \
50
- --adam_beta1 0.9 \
51
- --adam_beta2 0.98 \
52
- --mlm_probability 0.15 \
53
- --warmup_steps 10000 \
54
- --seed 42 \
55
- --preprocessing_num_workers 8 \
56
- --overwrite_output_dir True
57
- #--resume_from_checkpoint /root/NaN/dna-tokenizer/pretrain/models/model_len2_4096/checkpoint-36000
58
-
59
-
60
-
61
-
62
-
63
-
64
-
65
-
66
-
67
-
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/04.pretrain_200k.sh DELETED
@@ -1,62 +0,0 @@
1
- #!/bin/bash
2
- # Activate conda
3
- source ~/miniconda3/etc/profile.d/conda.sh
4
- conda activate bpe
5
-
6
- # select GPU
7
- export CUDA_VISIBLE_DEVICES=6
8
-
9
- work_dir=/home/n5huang/dna_token
10
- data_dir=${work_dir}/output_tokens
11
- cache_dir=${data_dir}/cache
12
-
13
- runnum=len2_5120
14
-
15
- # -p to use existing directory
16
- mkdir -p ${work_dir}/pretrain/models/${runnum}
17
- output_dir=${work_dir}/pretrain/models/${runnum}
18
-
19
- # CHANGED:
20
- # --config_name ${work_dir}/BPE_merge/config_5120.json -> --config_name ${work_dir}/pretrain/models/len2_5120/config.json
21
- # --tokenizer_name ${work_dir}/merge_bpe/... -> --tokenizer_name ${work_dir}/tokenizer_evaluation/merge_bpe/...
22
- # all tf-idf -> len2_5120 (model, tokenizer, data)
23
- # ADDED:
24
- # --resume_from_checkpoint /home/n5huang/dna_token/pretrain/models/len2_5120/checkpoint-100000
25
- # REMOVED:
26
- # overwrite_output_dir True
27
-
28
- cp /home/n5huang/dna-tokenizer/BPE_merge/04.pretrain_200k.sh ${output_dir}
29
- python /home/n5huang/dna-tokenizer/BPE_merge/run_mlm_200k.py \
30
- --output_dir ${output_dir} \
31
- --model_type bert \
32
- --tokenizer_name ${work_dir}/tokenizer_evaluation/merge_bpe/vocab_5120/merge_tokenizer_unigram_len2.json \
33
- --config_name ${work_dir}/pretrain/models/len2_5120/config.json \
34
- --project_name token_eval_len2_5120 \
35
- --do_train True \
36
- --model_max_length 512 \
37
- --max_seq_length 512 \
38
- --line_by_line True \
39
- --pad_to_max_length True \
40
- --train_file ${data_dir}/merge_bpe_5120_allchr_all_tokenized_train_chrOnly.tsv \
41
- --validation_file ${data_dir}/merge_bpe_5120_allchr_all_tokenized_val_chrOnly.tsv \
42
- --cache_dir ${cache_dir} \
43
- --use_fast_tokenizer True \
44
- --do_eval True \
45
- --gradient_accumulation_steps 1 \
46
- --per_device_train_batch_size 96 \
47
- --per_device_eval_batch_size 96 \
48
- --save_steps 1000 \
49
- --save_total_limit 10 \
50
- --max_steps 200000 \
51
- --logging_steps 1000 \
52
- --learning_rate 4e-5 \
53
- --adam_epsilon 1e-6 \
54
- --weight_decay 0.01 \
55
- --adam_beta1 0.9 \
56
- --adam_beta2 0.98 \
57
- --mlm_probability 0.15 \
58
- --warmup_steps 10000 \
59
- --seed 42 \
60
- --preprocessing_num_workers 8 \
61
- --resume_from_checkpoint /home/n5huang/dna_token/pretrain/models/len2_5120/checkpoint-100000
62
-
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/README.md DELETED
@@ -1,56 +0,0 @@
1
- ---
2
- tags:
3
- - generated_from_trainer
4
- metrics:
5
- - accuracy
6
- model-index:
7
- - name: len2_5120
8
- results: []
9
- ---
10
-
11
- <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
- should probably proofread and complete it, then remove this comment. -->
13
-
14
- # len2_5120
15
-
16
- This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
17
- It achieves the following results on the evaluation set:
18
- - Loss: 4.9009
19
- - Accuracy: 0.3043
20
-
21
- ## Model description
22
-
23
- More information needed
24
-
25
- ## Intended uses & limitations
26
-
27
- More information needed
28
-
29
- ## Training and evaluation data
30
-
31
- More information needed
32
-
33
- ## Training procedure
34
-
35
- ### Training hyperparameters
36
-
37
- The following hyperparameters were used during training:
38
- - learning_rate: 4e-05
39
- - train_batch_size: 96
40
- - eval_batch_size: 96
41
- - seed: 42
42
- - optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-06
43
- - lr_scheduler_type: linear
44
- - lr_scheduler_warmup_steps: 10000
45
- - training_steps: 200000
46
-
47
- ### Training results
48
-
49
-
50
-
51
- ### Framework versions
52
-
53
- - Transformers 4.35.2
54
- - Pytorch 2.8.0+cu128
55
- - Datasets 4.4.1
56
- - Tokenizers 0.15.2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/all_results.json DELETED
@@ -1,16 +0,0 @@
1
- {
2
- "epoch": 19.54,
3
- "eval_accuracy": 0.30433709022367145,
4
- "eval_loss": 4.900882244110107,
5
- "eval_runtime": 1259.8218,
6
- "eval_samples": 109173,
7
- "eval_samples_per_second": 86.657,
8
- "eval_step": 200000,
9
- "eval_steps_per_second": 0.903,
10
- "perplexity": 134.40830833002153,
11
- "train_loss": 2.48521353515625,
12
- "train_runtime": 258545.0504,
13
- "train_samples": 982503,
14
- "train_samples_per_second": 74.262,
15
- "train_steps_per_second": 0.774
16
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/checkpoint-191000/config.json DELETED
@@ -1,25 +0,0 @@
1
- {
2
- "architectures": [
3
- "BertForMaskedLM"
4
- ],
5
- "attention_probs_dropout_prob": 0.1,
6
- "classifier_dropout": null,
7
- "hidden_act": "gelu",
8
- "hidden_dropout_prob": 0.1,
9
- "hidden_size": 768,
10
- "initializer_range": 0.02,
11
- "intermediate_size": 3072,
12
- "layer_norm_eps": 1e-12,
13
- "max_length": 512,
14
- "max_position_embeddings": 512,
15
- "model_type": "bert",
16
- "num_attention_heads": 12,
17
- "num_hidden_layers": 12,
18
- "pad_token_id": 0,
19
- "position_embedding_type": "absolute",
20
- "torch_dtype": "float32",
21
- "transformers_version": "4.35.2",
22
- "type_vocab_size": 2,
23
- "use_cache": true,
24
- "vocab_size": 5120
25
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/checkpoint-191000/generation_config.json DELETED
@@ -1,6 +0,0 @@
1
- {
2
- "_from_model_config": true,
3
- "max_length": 512,
4
- "pad_token_id": 0,
5
- "transformers_version": "4.35.2"
6
- }
 
 
 
 
 
 
 
len2_5120/checkpoint-191000/model.safetensors DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:f3d088c80a86c3b8ceec56a90523c71b08b41b05395390ca7974259e7b86b46d
3
- size 359944256
 
 
 
 
len2_5120/checkpoint-191000/optimizer.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:ce8464e2e6a6ffe607fd3443edb93081a68f33e5002d9a3412e9dd405ea3e667
3
- size 720010699
 
 
 
 
len2_5120/checkpoint-191000/rng_state.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:74d64fa3c1245c787a712ad0a6d7fd62111d9933515a68c0b6ede8fcd6d12bbd
3
- size 14645
 
 
 
 
len2_5120/checkpoint-191000/scheduler.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:03801543841a49721ffa8d88e86d4edc246ccb0f1b28ddcb8c3e3197118fe676
3
- size 1465
 
 
 
 
len2_5120/checkpoint-191000/trainer_state.json DELETED
@@ -1,1165 +0,0 @@
1
- {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 18.661455788959454,
5
- "eval_steps": 500,
6
- "global_step": 191000,
7
- "is_hyper_param_search": false,
8
- "is_local_process_zero": true,
9
- "is_world_process_zero": true,
10
- "log_history": [
11
- {
12
- "epoch": 0.78,
13
- "learning_rate": 4.000000000000001e-06,
14
- "loss": 7.5832,
15
- "step": 1000
16
- },
17
- {
18
- "epoch": 1.56,
19
- "learning_rate": 8.000000000000001e-06,
20
- "loss": 6.9724,
21
- "step": 2000
22
- },
23
- {
24
- "epoch": 2.34,
25
- "learning_rate": 1.2e-05,
26
- "loss": 6.7695,
27
- "step": 3000
28
- },
29
- {
30
- "epoch": 3.12,
31
- "learning_rate": 1.6000000000000003e-05,
32
- "loss": 6.7002,
33
- "step": 4000
34
- },
35
- {
36
- "epoch": 3.91,
37
- "learning_rate": 2e-05,
38
- "loss": 6.666,
39
- "step": 5000
40
- },
41
- {
42
- "epoch": 4.69,
43
- "learning_rate": 2.4e-05,
44
- "loss": 6.6503,
45
- "step": 6000
46
- },
47
- {
48
- "epoch": 5.47,
49
- "learning_rate": 2.8e-05,
50
- "loss": 6.6407,
51
- "step": 7000
52
- },
53
- {
54
- "epoch": 6.25,
55
- "learning_rate": 3.2000000000000005e-05,
56
- "loss": 6.6277,
57
- "step": 8000
58
- },
59
- {
60
- "epoch": 7.03,
61
- "learning_rate": 3.6e-05,
62
- "loss": 6.6209,
63
- "step": 9000
64
- },
65
- {
66
- "epoch": 7.81,
67
- "learning_rate": 4e-05,
68
- "loss": 6.6135,
69
- "step": 10000
70
- },
71
- {
72
- "epoch": 8.59,
73
- "learning_rate": 3.955555555555556e-05,
74
- "loss": 6.6084,
75
- "step": 11000
76
- },
77
- {
78
- "epoch": 9.38,
79
- "learning_rate": 3.9111111111111115e-05,
80
- "loss": 6.6041,
81
- "step": 12000
82
- },
83
- {
84
- "epoch": 10.16,
85
- "learning_rate": 3.866666666666667e-05,
86
- "loss": 6.6013,
87
- "step": 13000
88
- },
89
- {
90
- "epoch": 10.94,
91
- "learning_rate": 3.8222222222222226e-05,
92
- "loss": 6.5994,
93
- "step": 14000
94
- },
95
- {
96
- "epoch": 11.72,
97
- "learning_rate": 3.777777777777778e-05,
98
- "loss": 6.594,
99
- "step": 15000
100
- },
101
- {
102
- "epoch": 12.5,
103
- "learning_rate": 3.733333333333334e-05,
104
- "loss": 6.5925,
105
- "step": 16000
106
- },
107
- {
108
- "epoch": 13.28,
109
- "learning_rate": 3.6888888888888896e-05,
110
- "loss": 6.5893,
111
- "step": 17000
112
- },
113
- {
114
- "epoch": 14.06,
115
- "learning_rate": 3.644444444444445e-05,
116
- "loss": 6.5875,
117
- "step": 18000
118
- },
119
- {
120
- "epoch": 14.84,
121
- "learning_rate": 3.6e-05,
122
- "loss": 6.5232,
123
- "step": 19000
124
- },
125
- {
126
- "epoch": 15.62,
127
- "learning_rate": 3.555555555555555e-05,
128
- "loss": 6.4358,
129
- "step": 20000
130
- },
131
- {
132
- "epoch": 16.41,
133
- "learning_rate": 3.511111111111111e-05,
134
- "loss": 6.3712,
135
- "step": 21000
136
- },
137
- {
138
- "epoch": 17.19,
139
- "learning_rate": 3.466666666666667e-05,
140
- "loss": 6.311,
141
- "step": 22000
142
- },
143
- {
144
- "epoch": 17.97,
145
- "learning_rate": 3.4222222222222224e-05,
146
- "loss": 6.2493,
147
- "step": 23000
148
- },
149
- {
150
- "epoch": 18.75,
151
- "learning_rate": 3.377777777777778e-05,
152
- "loss": 6.1597,
153
- "step": 24000
154
- },
155
- {
156
- "epoch": 19.53,
157
- "learning_rate": 3.3333333333333335e-05,
158
- "loss": 6.0276,
159
- "step": 25000
160
- },
161
- {
162
- "epoch": 20.31,
163
- "learning_rate": 3.288888888888889e-05,
164
- "loss": 5.8317,
165
- "step": 26000
166
- },
167
- {
168
- "epoch": 21.09,
169
- "learning_rate": 3.2444444444444446e-05,
170
- "loss": 5.6782,
171
- "step": 27000
172
- },
173
- {
174
- "epoch": 21.88,
175
- "learning_rate": 3.2000000000000005e-05,
176
- "loss": 5.5975,
177
- "step": 28000
178
- },
179
- {
180
- "epoch": 22.66,
181
- "learning_rate": 3.155555555555556e-05,
182
- "loss": 5.5369,
183
- "step": 29000
184
- },
185
- {
186
- "epoch": 23.44,
187
- "learning_rate": 3.111111111111112e-05,
188
- "loss": 5.4921,
189
- "step": 30000
190
- },
191
- {
192
- "epoch": 24.22,
193
- "learning_rate": 3.066666666666667e-05,
194
- "loss": 5.4554,
195
- "step": 31000
196
- },
197
- {
198
- "epoch": 25.0,
199
- "learning_rate": 3.0222222222222225e-05,
200
- "loss": 5.4237,
201
- "step": 32000
202
- },
203
- {
204
- "epoch": 25.78,
205
- "learning_rate": 2.977777777777778e-05,
206
- "loss": 5.3938,
207
- "step": 33000
208
- },
209
- {
210
- "epoch": 26.56,
211
- "learning_rate": 2.9333333333333333e-05,
212
- "loss": 5.3701,
213
- "step": 34000
214
- },
215
- {
216
- "epoch": 27.34,
217
- "learning_rate": 2.888888888888889e-05,
218
- "loss": 5.3452,
219
- "step": 35000
220
- },
221
- {
222
- "epoch": 28.12,
223
- "learning_rate": 2.8444444444444447e-05,
224
- "loss": 5.327,
225
- "step": 36000
226
- },
227
- {
228
- "epoch": 28.91,
229
- "learning_rate": 2.8e-05,
230
- "loss": 5.3063,
231
- "step": 37000
232
- },
233
- {
234
- "epoch": 29.69,
235
- "learning_rate": 2.755555555555556e-05,
236
- "loss": 5.2905,
237
- "step": 38000
238
- },
239
- {
240
- "epoch": 30.47,
241
- "learning_rate": 2.7111111111111114e-05,
242
- "loss": 5.2747,
243
- "step": 39000
244
- },
245
- {
246
- "epoch": 31.25,
247
- "learning_rate": 2.6666666666666667e-05,
248
- "loss": 5.2596,
249
- "step": 40000
250
- },
251
- {
252
- "epoch": 32.03,
253
- "learning_rate": 2.6222222222222226e-05,
254
- "loss": 5.2483,
255
- "step": 41000
256
- },
257
- {
258
- "epoch": 32.81,
259
- "learning_rate": 2.577777777777778e-05,
260
- "loss": 5.2336,
261
- "step": 42000
262
- },
263
- {
264
- "epoch": 33.59,
265
- "learning_rate": 2.5333333333333334e-05,
266
- "loss": 5.2214,
267
- "step": 43000
268
- },
269
- {
270
- "epoch": 34.38,
271
- "learning_rate": 2.4888888888888893e-05,
272
- "loss": 5.2103,
273
- "step": 44000
274
- },
275
- {
276
- "epoch": 35.16,
277
- "learning_rate": 2.444444444444445e-05,
278
- "loss": 5.2015,
279
- "step": 45000
280
- },
281
- {
282
- "epoch": 35.94,
283
- "learning_rate": 2.4e-05,
284
- "loss": 5.1907,
285
- "step": 46000
286
- },
287
- {
288
- "epoch": 36.72,
289
- "learning_rate": 2.355555555555556e-05,
290
- "loss": 5.1801,
291
- "step": 47000
292
- },
293
- {
294
- "epoch": 37.5,
295
- "learning_rate": 2.3111111111111112e-05,
296
- "loss": 5.1713,
297
- "step": 48000
298
- },
299
- {
300
- "epoch": 38.28,
301
- "learning_rate": 2.2666666666666668e-05,
302
- "loss": 5.1645,
303
- "step": 49000
304
- },
305
- {
306
- "epoch": 39.06,
307
- "learning_rate": 2.2222222222222227e-05,
308
- "loss": 5.1548,
309
- "step": 50000
310
- },
311
- {
312
- "epoch": 39.84,
313
- "learning_rate": 2.177777777777778e-05,
314
- "loss": 5.1462,
315
- "step": 51000
316
- },
317
- {
318
- "epoch": 40.62,
319
- "learning_rate": 2.1333333333333335e-05,
320
- "loss": 5.1417,
321
- "step": 52000
322
- },
323
- {
324
- "epoch": 41.41,
325
- "learning_rate": 2.088888888888889e-05,
326
- "loss": 5.1305,
327
- "step": 53000
328
- },
329
- {
330
- "epoch": 42.19,
331
- "learning_rate": 2.0444444444444446e-05,
332
- "loss": 5.1287,
333
- "step": 54000
334
- },
335
- {
336
- "epoch": 42.97,
337
- "learning_rate": 2e-05,
338
- "loss": 5.1202,
339
- "step": 55000
340
- },
341
- {
342
- "epoch": 43.75,
343
- "learning_rate": 1.9555555555555557e-05,
344
- "loss": 5.1139,
345
- "step": 56000
346
- },
347
- {
348
- "epoch": 44.53,
349
- "learning_rate": 1.9111111111111113e-05,
350
- "loss": 5.1081,
351
- "step": 57000
352
- },
353
- {
354
- "epoch": 45.31,
355
- "learning_rate": 1.866666666666667e-05,
356
- "loss": 5.1032,
357
- "step": 58000
358
- },
359
- {
360
- "epoch": 46.09,
361
- "learning_rate": 1.8222222222222224e-05,
362
- "loss": 5.0968,
363
- "step": 59000
364
- },
365
- {
366
- "epoch": 46.88,
367
- "learning_rate": 1.7777777777777777e-05,
368
- "loss": 5.0914,
369
- "step": 60000
370
- },
371
- {
372
- "epoch": 47.66,
373
- "learning_rate": 1.7333333333333336e-05,
374
- "loss": 5.085,
375
- "step": 61000
376
- },
377
- {
378
- "epoch": 48.44,
379
- "learning_rate": 1.688888888888889e-05,
380
- "loss": 5.0806,
381
- "step": 62000
382
- },
383
- {
384
- "epoch": 49.22,
385
- "learning_rate": 1.6444444444444444e-05,
386
- "loss": 5.0769,
387
- "step": 63000
388
- },
389
- {
390
- "epoch": 50.0,
391
- "learning_rate": 1.6000000000000003e-05,
392
- "loss": 5.0728,
393
- "step": 64000
394
- },
395
- {
396
- "epoch": 50.78,
397
- "learning_rate": 1.555555555555556e-05,
398
- "loss": 5.0692,
399
- "step": 65000
400
- },
401
- {
402
- "epoch": 51.56,
403
- "learning_rate": 1.5111111111111112e-05,
404
- "loss": 5.065,
405
- "step": 66000
406
- },
407
- {
408
- "epoch": 52.34,
409
- "learning_rate": 1.4666666666666666e-05,
410
- "loss": 5.0594,
411
- "step": 67000
412
- },
413
- {
414
- "epoch": 53.12,
415
- "learning_rate": 1.4222222222222224e-05,
416
- "loss": 5.0549,
417
- "step": 68000
418
- },
419
- {
420
- "epoch": 53.91,
421
- "learning_rate": 1.377777777777778e-05,
422
- "loss": 5.0538,
423
- "step": 69000
424
- },
425
- {
426
- "epoch": 54.69,
427
- "learning_rate": 1.3333333333333333e-05,
428
- "loss": 5.0482,
429
- "step": 70000
430
- },
431
- {
432
- "epoch": 55.47,
433
- "learning_rate": 1.288888888888889e-05,
434
- "loss": 5.0475,
435
- "step": 71000
436
- },
437
- {
438
- "epoch": 56.25,
439
- "learning_rate": 1.2444444444444446e-05,
440
- "loss": 5.0399,
441
- "step": 72000
442
- },
443
- {
444
- "epoch": 57.03,
445
- "learning_rate": 1.2e-05,
446
- "loss": 5.0406,
447
- "step": 73000
448
- },
449
- {
450
- "epoch": 57.81,
451
- "learning_rate": 1.1555555555555556e-05,
452
- "loss": 5.037,
453
- "step": 74000
454
- },
455
- {
456
- "epoch": 58.59,
457
- "learning_rate": 1.1111111111111113e-05,
458
- "loss": 5.0326,
459
- "step": 75000
460
- },
461
- {
462
- "epoch": 59.38,
463
- "learning_rate": 1.0666666666666667e-05,
464
- "loss": 5.031,
465
- "step": 76000
466
- },
467
- {
468
- "epoch": 60.16,
469
- "learning_rate": 1.0222222222222223e-05,
470
- "loss": 5.0291,
471
- "step": 77000
472
- },
473
- {
474
- "epoch": 60.94,
475
- "learning_rate": 9.777777777777779e-06,
476
- "loss": 5.0266,
477
- "step": 78000
478
- },
479
- {
480
- "epoch": 61.72,
481
- "learning_rate": 9.333333333333334e-06,
482
- "loss": 5.0235,
483
- "step": 79000
484
- },
485
- {
486
- "epoch": 62.5,
487
- "learning_rate": 8.888888888888888e-06,
488
- "loss": 5.0211,
489
- "step": 80000
490
- },
491
- {
492
- "epoch": 63.28,
493
- "learning_rate": 8.444444444444446e-06,
494
- "loss": 5.0214,
495
- "step": 81000
496
- },
497
- {
498
- "epoch": 64.06,
499
- "learning_rate": 8.000000000000001e-06,
500
- "loss": 5.0176,
501
- "step": 82000
502
- },
503
- {
504
- "epoch": 64.84,
505
- "learning_rate": 7.555555555555556e-06,
506
- "loss": 5.0157,
507
- "step": 83000
508
- },
509
- {
510
- "epoch": 65.62,
511
- "learning_rate": 7.111111111111112e-06,
512
- "loss": 5.014,
513
- "step": 84000
514
- },
515
- {
516
- "epoch": 66.41,
517
- "learning_rate": 6.666666666666667e-06,
518
- "loss": 5.0115,
519
- "step": 85000
520
- },
521
- {
522
- "epoch": 67.19,
523
- "learning_rate": 6.222222222222223e-06,
524
- "loss": 5.0087,
525
- "step": 86000
526
- },
527
- {
528
- "epoch": 67.97,
529
- "learning_rate": 5.777777777777778e-06,
530
- "loss": 5.0105,
531
- "step": 87000
532
- },
533
- {
534
- "epoch": 68.75,
535
- "learning_rate": 5.333333333333334e-06,
536
- "loss": 5.0062,
537
- "step": 88000
538
- },
539
- {
540
- "epoch": 69.53,
541
- "learning_rate": 4.888888888888889e-06,
542
- "loss": 5.0028,
543
- "step": 89000
544
- },
545
- {
546
- "epoch": 70.31,
547
- "learning_rate": 4.444444444444444e-06,
548
- "loss": 5.0066,
549
- "step": 90000
550
- },
551
- {
552
- "epoch": 71.09,
553
- "learning_rate": 4.000000000000001e-06,
554
- "loss": 5.0032,
555
- "step": 91000
556
- },
557
- {
558
- "epoch": 71.88,
559
- "learning_rate": 3.555555555555556e-06,
560
- "loss": 5.0022,
561
- "step": 92000
562
- },
563
- {
564
- "epoch": 72.66,
565
- "learning_rate": 3.1111111111111116e-06,
566
- "loss": 5.0024,
567
- "step": 93000
568
- },
569
- {
570
- "epoch": 73.44,
571
- "learning_rate": 2.666666666666667e-06,
572
- "loss": 5.0007,
573
- "step": 94000
574
- },
575
- {
576
- "epoch": 74.22,
577
- "learning_rate": 2.222222222222222e-06,
578
- "loss": 5.0008,
579
- "step": 95000
580
- },
581
- {
582
- "epoch": 75.0,
583
- "learning_rate": 1.777777777777778e-06,
584
- "loss": 4.9978,
585
- "step": 96000
586
- },
587
- {
588
- "epoch": 75.78,
589
- "learning_rate": 1.3333333333333334e-06,
590
- "loss": 4.9966,
591
- "step": 97000
592
- },
593
- {
594
- "epoch": 76.56,
595
- "learning_rate": 8.88888888888889e-07,
596
- "loss": 4.9994,
597
- "step": 98000
598
- },
599
- {
600
- "epoch": 77.34,
601
- "learning_rate": 4.444444444444445e-07,
602
- "loss": 4.9966,
603
- "step": 99000
604
- },
605
- {
606
- "epoch": 78.12,
607
- "learning_rate": 0.0,
608
- "loss": 4.9976,
609
- "step": 100000
610
- },
611
- {
612
- "epoch": 9.87,
613
- "learning_rate": 2.0842105263157895e-05,
614
- "loss": 5.0155,
615
- "step": 101000
616
- },
617
- {
618
- "epoch": 9.97,
619
- "learning_rate": 2.0631578947368425e-05,
620
- "loss": 5.024,
621
- "step": 102000
622
- },
623
- {
624
- "epoch": 10.06,
625
- "learning_rate": 2.0421052631578952e-05,
626
- "loss": 5.0241,
627
- "step": 103000
628
- },
629
- {
630
- "epoch": 10.16,
631
- "learning_rate": 2.0210526315789475e-05,
632
- "loss": 5.0218,
633
- "step": 104000
634
- },
635
- {
636
- "epoch": 10.26,
637
- "learning_rate": 2e-05,
638
- "loss": 5.0231,
639
- "step": 105000
640
- },
641
- {
642
- "epoch": 10.36,
643
- "learning_rate": 1.9789473684210528e-05,
644
- "loss": 5.0231,
645
- "step": 106000
646
- },
647
- {
648
- "epoch": 10.45,
649
- "learning_rate": 1.9578947368421055e-05,
650
- "loss": 5.0198,
651
- "step": 107000
652
- },
653
- {
654
- "epoch": 10.55,
655
- "learning_rate": 1.936842105263158e-05,
656
- "loss": 5.0158,
657
- "step": 108000
658
- },
659
- {
660
- "epoch": 10.65,
661
- "learning_rate": 1.9157894736842108e-05,
662
- "loss": 5.0122,
663
- "step": 109000
664
- },
665
- {
666
- "epoch": 10.75,
667
- "learning_rate": 1.894736842105263e-05,
668
- "loss": 5.0165,
669
- "step": 110000
670
- },
671
- {
672
- "epoch": 10.85,
673
- "learning_rate": 1.873684210526316e-05,
674
- "loss": 5.0183,
675
- "step": 111000
676
- },
677
- {
678
- "epoch": 10.94,
679
- "learning_rate": 1.8526315789473684e-05,
680
- "loss": 5.01,
681
- "step": 112000
682
- },
683
- {
684
- "epoch": 11.04,
685
- "learning_rate": 1.831578947368421e-05,
686
- "loss": 5.007,
687
- "step": 113000
688
- },
689
- {
690
- "epoch": 11.14,
691
- "learning_rate": 1.810526315789474e-05,
692
- "loss": 5.0059,
693
- "step": 114000
694
- },
695
- {
696
- "epoch": 11.24,
697
- "learning_rate": 1.7894736842105264e-05,
698
- "loss": 5.0089,
699
- "step": 115000
700
- },
701
- {
702
- "epoch": 11.33,
703
- "learning_rate": 1.768421052631579e-05,
704
- "loss": 5.0081,
705
- "step": 116000
706
- },
707
- {
708
- "epoch": 11.43,
709
- "learning_rate": 1.7473684210526317e-05,
710
- "loss": 5.002,
711
- "step": 117000
712
- },
713
- {
714
- "epoch": 11.53,
715
- "learning_rate": 1.7263157894736843e-05,
716
- "loss": 5.0052,
717
- "step": 118000
718
- },
719
- {
720
- "epoch": 11.63,
721
- "learning_rate": 1.705263157894737e-05,
722
- "loss": 5.0079,
723
- "step": 119000
724
- },
725
- {
726
- "epoch": 11.72,
727
- "learning_rate": 1.6842105263157896e-05,
728
- "loss": 4.9985,
729
- "step": 120000
730
- },
731
- {
732
- "epoch": 11.82,
733
- "learning_rate": 1.6631578947368423e-05,
734
- "loss": 5.006,
735
- "step": 121000
736
- },
737
- {
738
- "epoch": 11.92,
739
- "learning_rate": 1.642105263157895e-05,
740
- "loss": 4.9956,
741
- "step": 122000
742
- },
743
- {
744
- "epoch": 12.02,
745
- "learning_rate": 1.6210526315789473e-05,
746
- "loss": 4.9973,
747
- "step": 123000
748
- },
749
- {
750
- "epoch": 12.12,
751
- "learning_rate": 1.6000000000000003e-05,
752
- "loss": 4.9976,
753
- "step": 124000
754
- },
755
- {
756
- "epoch": 12.21,
757
- "learning_rate": 1.578947368421053e-05,
758
- "loss": 4.9882,
759
- "step": 125000
760
- },
761
- {
762
- "epoch": 12.31,
763
- "learning_rate": 1.5578947368421052e-05,
764
- "loss": 4.9908,
765
- "step": 126000
766
- },
767
- {
768
- "epoch": 12.41,
769
- "learning_rate": 1.536842105263158e-05,
770
- "loss": 5.0038,
771
- "step": 127000
772
- },
773
- {
774
- "epoch": 12.51,
775
- "learning_rate": 1.5157894736842107e-05,
776
- "loss": 4.9923,
777
- "step": 128000
778
- },
779
- {
780
- "epoch": 12.6,
781
- "learning_rate": 1.4947368421052632e-05,
782
- "loss": 4.9847,
783
- "step": 129000
784
- },
785
- {
786
- "epoch": 12.7,
787
- "learning_rate": 1.4736842105263159e-05,
788
- "loss": 4.9864,
789
- "step": 130000
790
- },
791
- {
792
- "epoch": 12.8,
793
- "learning_rate": 1.4526315789473687e-05,
794
- "loss": 4.9903,
795
- "step": 131000
796
- },
797
- {
798
- "epoch": 12.9,
799
- "learning_rate": 1.4315789473684212e-05,
800
- "loss": 4.9812,
801
- "step": 132000
802
- },
803
- {
804
- "epoch": 12.99,
805
- "learning_rate": 1.4105263157894738e-05,
806
- "loss": 4.9923,
807
- "step": 133000
808
- },
809
- {
810
- "epoch": 13.09,
811
- "learning_rate": 1.3894736842105265e-05,
812
- "loss": 4.9838,
813
- "step": 134000
814
- },
815
- {
816
- "epoch": 13.19,
817
- "learning_rate": 1.3684210526315791e-05,
818
- "loss": 4.977,
819
- "step": 135000
820
- },
821
- {
822
- "epoch": 13.29,
823
- "learning_rate": 1.3473684210526316e-05,
824
- "loss": 4.9743,
825
- "step": 136000
826
- },
827
- {
828
- "epoch": 13.39,
829
- "learning_rate": 1.3263157894736843e-05,
830
- "loss": 4.9797,
831
- "step": 137000
832
- },
833
- {
834
- "epoch": 13.48,
835
- "learning_rate": 1.305263157894737e-05,
836
- "loss": 4.9915,
837
- "step": 138000
838
- },
839
- {
840
- "epoch": 13.58,
841
- "learning_rate": 1.2842105263157896e-05,
842
- "loss": 4.9762,
843
- "step": 139000
844
- },
845
- {
846
- "epoch": 13.68,
847
- "learning_rate": 1.263157894736842e-05,
848
- "loss": 4.9771,
849
- "step": 140000
850
- },
851
- {
852
- "epoch": 13.78,
853
- "learning_rate": 1.2421052631578949e-05,
854
- "loss": 4.9763,
855
- "step": 141000
856
- },
857
- {
858
- "epoch": 13.87,
859
- "learning_rate": 1.2210526315789475e-05,
860
- "loss": 4.9706,
861
- "step": 142000
862
- },
863
- {
864
- "epoch": 13.97,
865
- "learning_rate": 1.2e-05,
866
- "loss": 4.9731,
867
- "step": 143000
868
- },
869
- {
870
- "epoch": 14.07,
871
- "learning_rate": 1.1789473684210527e-05,
872
- "loss": 4.9736,
873
- "step": 144000
874
- },
875
- {
876
- "epoch": 14.17,
877
- "learning_rate": 1.1578947368421053e-05,
878
- "loss": 4.9704,
879
- "step": 145000
880
- },
881
- {
882
- "epoch": 14.26,
883
- "learning_rate": 1.136842105263158e-05,
884
- "loss": 4.9762,
885
- "step": 146000
886
- },
887
- {
888
- "epoch": 14.36,
889
- "learning_rate": 1.1157894736842105e-05,
890
- "loss": 4.9709,
891
- "step": 147000
892
- },
893
- {
894
- "epoch": 14.46,
895
- "learning_rate": 1.0947368421052633e-05,
896
- "loss": 4.957,
897
- "step": 148000
898
- },
899
- {
900
- "epoch": 14.56,
901
- "learning_rate": 1.073684210526316e-05,
902
- "loss": 4.9666,
903
- "step": 149000
904
- },
905
- {
906
- "epoch": 14.66,
907
- "learning_rate": 1.0526315789473684e-05,
908
- "loss": 4.9623,
909
- "step": 150000
910
- },
911
- {
912
- "epoch": 14.75,
913
- "learning_rate": 1.0315789473684213e-05,
914
- "loss": 4.9613,
915
- "step": 151000
916
- },
917
- {
918
- "epoch": 14.85,
919
- "learning_rate": 1.0105263157894738e-05,
920
- "loss": 4.9642,
921
- "step": 152000
922
- },
923
- {
924
- "epoch": 14.95,
925
- "learning_rate": 9.894736842105264e-06,
926
- "loss": 4.9605,
927
- "step": 153000
928
- },
929
- {
930
- "epoch": 15.05,
931
- "learning_rate": 9.68421052631579e-06,
932
- "loss": 4.9688,
933
- "step": 154000
934
- },
935
- {
936
- "epoch": 15.14,
937
- "learning_rate": 9.473684210526315e-06,
938
- "loss": 4.9635,
939
- "step": 155000
940
- },
941
- {
942
- "epoch": 15.24,
943
- "learning_rate": 9.263157894736842e-06,
944
- "loss": 4.9587,
945
- "step": 156000
946
- },
947
- {
948
- "epoch": 15.34,
949
- "learning_rate": 9.05263157894737e-06,
950
- "loss": 4.9633,
951
- "step": 157000
952
- },
953
- {
954
- "epoch": 15.44,
955
- "learning_rate": 8.842105263157895e-06,
956
- "loss": 4.9613,
957
- "step": 158000
958
- },
959
- {
960
- "epoch": 15.53,
961
- "learning_rate": 8.631578947368422e-06,
962
- "loss": 4.9629,
963
- "step": 159000
964
- },
965
- {
966
- "epoch": 15.63,
967
- "learning_rate": 8.421052631578948e-06,
968
- "loss": 4.9547,
969
- "step": 160000
970
- },
971
- {
972
- "epoch": 15.73,
973
- "learning_rate": 8.210526315789475e-06,
974
- "loss": 4.9541,
975
- "step": 161000
976
- },
977
- {
978
- "epoch": 15.83,
979
- "learning_rate": 8.000000000000001e-06,
980
- "loss": 4.9532,
981
- "step": 162000
982
- },
983
- {
984
- "epoch": 15.93,
985
- "learning_rate": 7.789473684210526e-06,
986
- "loss": 4.9461,
987
- "step": 163000
988
- },
989
- {
990
- "epoch": 16.02,
991
- "learning_rate": 7.578947368421054e-06,
992
- "loss": 4.9518,
993
- "step": 164000
994
- },
995
- {
996
- "epoch": 16.12,
997
- "learning_rate": 7.368421052631579e-06,
998
- "loss": 4.9543,
999
- "step": 165000
1000
- },
1001
- {
1002
- "epoch": 16.22,
1003
- "learning_rate": 7.157894736842106e-06,
1004
- "loss": 4.9526,
1005
- "step": 166000
1006
- },
1007
- {
1008
- "epoch": 16.32,
1009
- "learning_rate": 6.947368421052632e-06,
1010
- "loss": 4.9447,
1011
- "step": 167000
1012
- },
1013
- {
1014
- "epoch": 16.41,
1015
- "learning_rate": 6.736842105263158e-06,
1016
- "loss": 4.9545,
1017
- "step": 168000
1018
- },
1019
- {
1020
- "epoch": 16.51,
1021
- "learning_rate": 6.526315789473685e-06,
1022
- "loss": 4.9448,
1023
- "step": 169000
1024
- },
1025
- {
1026
- "epoch": 16.61,
1027
- "learning_rate": 6.31578947368421e-06,
1028
- "loss": 4.9521,
1029
- "step": 170000
1030
- },
1031
- {
1032
- "epoch": 16.71,
1033
- "learning_rate": 6.105263157894738e-06,
1034
- "loss": 4.9503,
1035
- "step": 171000
1036
- },
1037
- {
1038
- "epoch": 16.81,
1039
- "learning_rate": 5.8947368421052634e-06,
1040
- "loss": 4.9537,
1041
- "step": 172000
1042
- },
1043
- {
1044
- "epoch": 16.9,
1045
- "learning_rate": 5.68421052631579e-06,
1046
- "loss": 4.9446,
1047
- "step": 173000
1048
- },
1049
- {
1050
- "epoch": 17.0,
1051
- "learning_rate": 5.4736842105263165e-06,
1052
- "loss": 4.9397,
1053
- "step": 174000
1054
- },
1055
- {
1056
- "epoch": 17.1,
1057
- "learning_rate": 5.263157894736842e-06,
1058
- "loss": 4.9541,
1059
- "step": 175000
1060
- },
1061
- {
1062
- "epoch": 17.2,
1063
- "learning_rate": 5.052631578947369e-06,
1064
- "loss": 4.9449,
1065
- "step": 176000
1066
- },
1067
- {
1068
- "epoch": 17.29,
1069
- "learning_rate": 4.842105263157895e-06,
1070
- "loss": 4.9447,
1071
- "step": 177000
1072
- },
1073
- {
1074
- "epoch": 17.39,
1075
- "learning_rate": 4.631578947368421e-06,
1076
- "loss": 4.9344,
1077
- "step": 178000
1078
- },
1079
- {
1080
- "epoch": 17.49,
1081
- "learning_rate": 4.4210526315789476e-06,
1082
- "loss": 4.933,
1083
- "step": 179000
1084
- },
1085
- {
1086
- "epoch": 17.59,
1087
- "learning_rate": 4.210526315789474e-06,
1088
- "loss": 4.9397,
1089
- "step": 180000
1090
- },
1091
- {
1092
- "epoch": 17.68,
1093
- "learning_rate": 4.000000000000001e-06,
1094
- "loss": 4.9431,
1095
- "step": 181000
1096
- },
1097
- {
1098
- "epoch": 17.78,
1099
- "learning_rate": 3.789473684210527e-06,
1100
- "loss": 4.9378,
1101
- "step": 182000
1102
- },
1103
- {
1104
- "epoch": 17.88,
1105
- "learning_rate": 3.578947368421053e-06,
1106
- "loss": 4.9399,
1107
- "step": 183000
1108
- },
1109
- {
1110
- "epoch": 17.98,
1111
- "learning_rate": 3.368421052631579e-06,
1112
- "loss": 4.9481,
1113
- "step": 184000
1114
- },
1115
- {
1116
- "epoch": 18.08,
1117
- "learning_rate": 3.157894736842105e-06,
1118
- "loss": 4.9399,
1119
- "step": 185000
1120
- },
1121
- {
1122
- "epoch": 18.17,
1123
- "learning_rate": 2.9473684210526317e-06,
1124
- "loss": 4.9365,
1125
- "step": 186000
1126
- },
1127
- {
1128
- "epoch": 18.27,
1129
- "learning_rate": 2.7368421052631583e-06,
1130
- "loss": 4.9307,
1131
- "step": 187000
1132
- },
1133
- {
1134
- "epoch": 18.37,
1135
- "learning_rate": 2.5263157894736844e-06,
1136
- "loss": 4.9391,
1137
- "step": 188000
1138
- },
1139
- {
1140
- "epoch": 18.47,
1141
- "learning_rate": 2.3157894736842105e-06,
1142
- "loss": 4.9331,
1143
- "step": 189000
1144
- },
1145
- {
1146
- "epoch": 18.56,
1147
- "learning_rate": 2.105263157894737e-06,
1148
- "loss": 4.948,
1149
- "step": 190000
1150
- },
1151
- {
1152
- "epoch": 18.66,
1153
- "learning_rate": 1.8947368421052634e-06,
1154
- "loss": 4.9407,
1155
- "step": 191000
1156
- }
1157
- ],
1158
- "logging_steps": 1000,
1159
- "max_steps": 200000,
1160
- "num_train_epochs": 20,
1161
- "save_steps": 1000,
1162
- "total_flos": 2.250667778458596e+19,
1163
- "trial_name": null,
1164
- "trial_params": null
1165
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/checkpoint-191000/training_args.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:c9ff5d6388a59708cb6305a7fab984d438e385b1835680f9629b90c23c5cf704
3
- size 5137
 
 
 
 
len2_5120/checkpoint-192000/config.json DELETED
@@ -1,25 +0,0 @@
1
- {
2
- "architectures": [
3
- "BertForMaskedLM"
4
- ],
5
- "attention_probs_dropout_prob": 0.1,
6
- "classifier_dropout": null,
7
- "hidden_act": "gelu",
8
- "hidden_dropout_prob": 0.1,
9
- "hidden_size": 768,
10
- "initializer_range": 0.02,
11
- "intermediate_size": 3072,
12
- "layer_norm_eps": 1e-12,
13
- "max_length": 512,
14
- "max_position_embeddings": 512,
15
- "model_type": "bert",
16
- "num_attention_heads": 12,
17
- "num_hidden_layers": 12,
18
- "pad_token_id": 0,
19
- "position_embedding_type": "absolute",
20
- "torch_dtype": "float32",
21
- "transformers_version": "4.35.2",
22
- "type_vocab_size": 2,
23
- "use_cache": true,
24
- "vocab_size": 5120
25
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/checkpoint-192000/generation_config.json DELETED
@@ -1,6 +0,0 @@
1
- {
2
- "_from_model_config": true,
3
- "max_length": 512,
4
- "pad_token_id": 0,
5
- "transformers_version": "4.35.2"
6
- }
 
 
 
 
 
 
 
len2_5120/checkpoint-192000/model.safetensors DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:4cbffd267340e7a09c8e1143725cb945c07c0c1658055e0fba16d5ae1b14c7ac
3
- size 359944256
 
 
 
 
len2_5120/checkpoint-192000/optimizer.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:06d737a339b674a47bbb393421311b07894bc8deb11a35745119f729e10925b8
3
- size 720010699
 
 
 
 
len2_5120/checkpoint-192000/rng_state.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:94f051d28f4c9a4e5765ca63afa821d32882244885368ce9c7547bfc4cd15fb2
3
- size 14645
 
 
 
 
len2_5120/checkpoint-192000/scheduler.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:8987c093fe567c44cd8f1b73e4b15e5d4b9c5495e847cc1b11a0915470ccf64f
3
- size 1465
 
 
 
 
len2_5120/checkpoint-192000/trainer_state.json DELETED
@@ -1,1171 +0,0 @@
1
- {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 18.75915974596971,
5
- "eval_steps": 500,
6
- "global_step": 192000,
7
- "is_hyper_param_search": false,
8
- "is_local_process_zero": true,
9
- "is_world_process_zero": true,
10
- "log_history": [
11
- {
12
- "epoch": 0.78,
13
- "learning_rate": 4.000000000000001e-06,
14
- "loss": 7.5832,
15
- "step": 1000
16
- },
17
- {
18
- "epoch": 1.56,
19
- "learning_rate": 8.000000000000001e-06,
20
- "loss": 6.9724,
21
- "step": 2000
22
- },
23
- {
24
- "epoch": 2.34,
25
- "learning_rate": 1.2e-05,
26
- "loss": 6.7695,
27
- "step": 3000
28
- },
29
- {
30
- "epoch": 3.12,
31
- "learning_rate": 1.6000000000000003e-05,
32
- "loss": 6.7002,
33
- "step": 4000
34
- },
35
- {
36
- "epoch": 3.91,
37
- "learning_rate": 2e-05,
38
- "loss": 6.666,
39
- "step": 5000
40
- },
41
- {
42
- "epoch": 4.69,
43
- "learning_rate": 2.4e-05,
44
- "loss": 6.6503,
45
- "step": 6000
46
- },
47
- {
48
- "epoch": 5.47,
49
- "learning_rate": 2.8e-05,
50
- "loss": 6.6407,
51
- "step": 7000
52
- },
53
- {
54
- "epoch": 6.25,
55
- "learning_rate": 3.2000000000000005e-05,
56
- "loss": 6.6277,
57
- "step": 8000
58
- },
59
- {
60
- "epoch": 7.03,
61
- "learning_rate": 3.6e-05,
62
- "loss": 6.6209,
63
- "step": 9000
64
- },
65
- {
66
- "epoch": 7.81,
67
- "learning_rate": 4e-05,
68
- "loss": 6.6135,
69
- "step": 10000
70
- },
71
- {
72
- "epoch": 8.59,
73
- "learning_rate": 3.955555555555556e-05,
74
- "loss": 6.6084,
75
- "step": 11000
76
- },
77
- {
78
- "epoch": 9.38,
79
- "learning_rate": 3.9111111111111115e-05,
80
- "loss": 6.6041,
81
- "step": 12000
82
- },
83
- {
84
- "epoch": 10.16,
85
- "learning_rate": 3.866666666666667e-05,
86
- "loss": 6.6013,
87
- "step": 13000
88
- },
89
- {
90
- "epoch": 10.94,
91
- "learning_rate": 3.8222222222222226e-05,
92
- "loss": 6.5994,
93
- "step": 14000
94
- },
95
- {
96
- "epoch": 11.72,
97
- "learning_rate": 3.777777777777778e-05,
98
- "loss": 6.594,
99
- "step": 15000
100
- },
101
- {
102
- "epoch": 12.5,
103
- "learning_rate": 3.733333333333334e-05,
104
- "loss": 6.5925,
105
- "step": 16000
106
- },
107
- {
108
- "epoch": 13.28,
109
- "learning_rate": 3.6888888888888896e-05,
110
- "loss": 6.5893,
111
- "step": 17000
112
- },
113
- {
114
- "epoch": 14.06,
115
- "learning_rate": 3.644444444444445e-05,
116
- "loss": 6.5875,
117
- "step": 18000
118
- },
119
- {
120
- "epoch": 14.84,
121
- "learning_rate": 3.6e-05,
122
- "loss": 6.5232,
123
- "step": 19000
124
- },
125
- {
126
- "epoch": 15.62,
127
- "learning_rate": 3.555555555555555e-05,
128
- "loss": 6.4358,
129
- "step": 20000
130
- },
131
- {
132
- "epoch": 16.41,
133
- "learning_rate": 3.511111111111111e-05,
134
- "loss": 6.3712,
135
- "step": 21000
136
- },
137
- {
138
- "epoch": 17.19,
139
- "learning_rate": 3.466666666666667e-05,
140
- "loss": 6.311,
141
- "step": 22000
142
- },
143
- {
144
- "epoch": 17.97,
145
- "learning_rate": 3.4222222222222224e-05,
146
- "loss": 6.2493,
147
- "step": 23000
148
- },
149
- {
150
- "epoch": 18.75,
151
- "learning_rate": 3.377777777777778e-05,
152
- "loss": 6.1597,
153
- "step": 24000
154
- },
155
- {
156
- "epoch": 19.53,
157
- "learning_rate": 3.3333333333333335e-05,
158
- "loss": 6.0276,
159
- "step": 25000
160
- },
161
- {
162
- "epoch": 20.31,
163
- "learning_rate": 3.288888888888889e-05,
164
- "loss": 5.8317,
165
- "step": 26000
166
- },
167
- {
168
- "epoch": 21.09,
169
- "learning_rate": 3.2444444444444446e-05,
170
- "loss": 5.6782,
171
- "step": 27000
172
- },
173
- {
174
- "epoch": 21.88,
175
- "learning_rate": 3.2000000000000005e-05,
176
- "loss": 5.5975,
177
- "step": 28000
178
- },
179
- {
180
- "epoch": 22.66,
181
- "learning_rate": 3.155555555555556e-05,
182
- "loss": 5.5369,
183
- "step": 29000
184
- },
185
- {
186
- "epoch": 23.44,
187
- "learning_rate": 3.111111111111112e-05,
188
- "loss": 5.4921,
189
- "step": 30000
190
- },
191
- {
192
- "epoch": 24.22,
193
- "learning_rate": 3.066666666666667e-05,
194
- "loss": 5.4554,
195
- "step": 31000
196
- },
197
- {
198
- "epoch": 25.0,
199
- "learning_rate": 3.0222222222222225e-05,
200
- "loss": 5.4237,
201
- "step": 32000
202
- },
203
- {
204
- "epoch": 25.78,
205
- "learning_rate": 2.977777777777778e-05,
206
- "loss": 5.3938,
207
- "step": 33000
208
- },
209
- {
210
- "epoch": 26.56,
211
- "learning_rate": 2.9333333333333333e-05,
212
- "loss": 5.3701,
213
- "step": 34000
214
- },
215
- {
216
- "epoch": 27.34,
217
- "learning_rate": 2.888888888888889e-05,
218
- "loss": 5.3452,
219
- "step": 35000
220
- },
221
- {
222
- "epoch": 28.12,
223
- "learning_rate": 2.8444444444444447e-05,
224
- "loss": 5.327,
225
- "step": 36000
226
- },
227
- {
228
- "epoch": 28.91,
229
- "learning_rate": 2.8e-05,
230
- "loss": 5.3063,
231
- "step": 37000
232
- },
233
- {
234
- "epoch": 29.69,
235
- "learning_rate": 2.755555555555556e-05,
236
- "loss": 5.2905,
237
- "step": 38000
238
- },
239
- {
240
- "epoch": 30.47,
241
- "learning_rate": 2.7111111111111114e-05,
242
- "loss": 5.2747,
243
- "step": 39000
244
- },
245
- {
246
- "epoch": 31.25,
247
- "learning_rate": 2.6666666666666667e-05,
248
- "loss": 5.2596,
249
- "step": 40000
250
- },
251
- {
252
- "epoch": 32.03,
253
- "learning_rate": 2.6222222222222226e-05,
254
- "loss": 5.2483,
255
- "step": 41000
256
- },
257
- {
258
- "epoch": 32.81,
259
- "learning_rate": 2.577777777777778e-05,
260
- "loss": 5.2336,
261
- "step": 42000
262
- },
263
- {
264
- "epoch": 33.59,
265
- "learning_rate": 2.5333333333333334e-05,
266
- "loss": 5.2214,
267
- "step": 43000
268
- },
269
- {
270
- "epoch": 34.38,
271
- "learning_rate": 2.4888888888888893e-05,
272
- "loss": 5.2103,
273
- "step": 44000
274
- },
275
- {
276
- "epoch": 35.16,
277
- "learning_rate": 2.444444444444445e-05,
278
- "loss": 5.2015,
279
- "step": 45000
280
- },
281
- {
282
- "epoch": 35.94,
283
- "learning_rate": 2.4e-05,
284
- "loss": 5.1907,
285
- "step": 46000
286
- },
287
- {
288
- "epoch": 36.72,
289
- "learning_rate": 2.355555555555556e-05,
290
- "loss": 5.1801,
291
- "step": 47000
292
- },
293
- {
294
- "epoch": 37.5,
295
- "learning_rate": 2.3111111111111112e-05,
296
- "loss": 5.1713,
297
- "step": 48000
298
- },
299
- {
300
- "epoch": 38.28,
301
- "learning_rate": 2.2666666666666668e-05,
302
- "loss": 5.1645,
303
- "step": 49000
304
- },
305
- {
306
- "epoch": 39.06,
307
- "learning_rate": 2.2222222222222227e-05,
308
- "loss": 5.1548,
309
- "step": 50000
310
- },
311
- {
312
- "epoch": 39.84,
313
- "learning_rate": 2.177777777777778e-05,
314
- "loss": 5.1462,
315
- "step": 51000
316
- },
317
- {
318
- "epoch": 40.62,
319
- "learning_rate": 2.1333333333333335e-05,
320
- "loss": 5.1417,
321
- "step": 52000
322
- },
323
- {
324
- "epoch": 41.41,
325
- "learning_rate": 2.088888888888889e-05,
326
- "loss": 5.1305,
327
- "step": 53000
328
- },
329
- {
330
- "epoch": 42.19,
331
- "learning_rate": 2.0444444444444446e-05,
332
- "loss": 5.1287,
333
- "step": 54000
334
- },
335
- {
336
- "epoch": 42.97,
337
- "learning_rate": 2e-05,
338
- "loss": 5.1202,
339
- "step": 55000
340
- },
341
- {
342
- "epoch": 43.75,
343
- "learning_rate": 1.9555555555555557e-05,
344
- "loss": 5.1139,
345
- "step": 56000
346
- },
347
- {
348
- "epoch": 44.53,
349
- "learning_rate": 1.9111111111111113e-05,
350
- "loss": 5.1081,
351
- "step": 57000
352
- },
353
- {
354
- "epoch": 45.31,
355
- "learning_rate": 1.866666666666667e-05,
356
- "loss": 5.1032,
357
- "step": 58000
358
- },
359
- {
360
- "epoch": 46.09,
361
- "learning_rate": 1.8222222222222224e-05,
362
- "loss": 5.0968,
363
- "step": 59000
364
- },
365
- {
366
- "epoch": 46.88,
367
- "learning_rate": 1.7777777777777777e-05,
368
- "loss": 5.0914,
369
- "step": 60000
370
- },
371
- {
372
- "epoch": 47.66,
373
- "learning_rate": 1.7333333333333336e-05,
374
- "loss": 5.085,
375
- "step": 61000
376
- },
377
- {
378
- "epoch": 48.44,
379
- "learning_rate": 1.688888888888889e-05,
380
- "loss": 5.0806,
381
- "step": 62000
382
- },
383
- {
384
- "epoch": 49.22,
385
- "learning_rate": 1.6444444444444444e-05,
386
- "loss": 5.0769,
387
- "step": 63000
388
- },
389
- {
390
- "epoch": 50.0,
391
- "learning_rate": 1.6000000000000003e-05,
392
- "loss": 5.0728,
393
- "step": 64000
394
- },
395
- {
396
- "epoch": 50.78,
397
- "learning_rate": 1.555555555555556e-05,
398
- "loss": 5.0692,
399
- "step": 65000
400
- },
401
- {
402
- "epoch": 51.56,
403
- "learning_rate": 1.5111111111111112e-05,
404
- "loss": 5.065,
405
- "step": 66000
406
- },
407
- {
408
- "epoch": 52.34,
409
- "learning_rate": 1.4666666666666666e-05,
410
- "loss": 5.0594,
411
- "step": 67000
412
- },
413
- {
414
- "epoch": 53.12,
415
- "learning_rate": 1.4222222222222224e-05,
416
- "loss": 5.0549,
417
- "step": 68000
418
- },
419
- {
420
- "epoch": 53.91,
421
- "learning_rate": 1.377777777777778e-05,
422
- "loss": 5.0538,
423
- "step": 69000
424
- },
425
- {
426
- "epoch": 54.69,
427
- "learning_rate": 1.3333333333333333e-05,
428
- "loss": 5.0482,
429
- "step": 70000
430
- },
431
- {
432
- "epoch": 55.47,
433
- "learning_rate": 1.288888888888889e-05,
434
- "loss": 5.0475,
435
- "step": 71000
436
- },
437
- {
438
- "epoch": 56.25,
439
- "learning_rate": 1.2444444444444446e-05,
440
- "loss": 5.0399,
441
- "step": 72000
442
- },
443
- {
444
- "epoch": 57.03,
445
- "learning_rate": 1.2e-05,
446
- "loss": 5.0406,
447
- "step": 73000
448
- },
449
- {
450
- "epoch": 57.81,
451
- "learning_rate": 1.1555555555555556e-05,
452
- "loss": 5.037,
453
- "step": 74000
454
- },
455
- {
456
- "epoch": 58.59,
457
- "learning_rate": 1.1111111111111113e-05,
458
- "loss": 5.0326,
459
- "step": 75000
460
- },
461
- {
462
- "epoch": 59.38,
463
- "learning_rate": 1.0666666666666667e-05,
464
- "loss": 5.031,
465
- "step": 76000
466
- },
467
- {
468
- "epoch": 60.16,
469
- "learning_rate": 1.0222222222222223e-05,
470
- "loss": 5.0291,
471
- "step": 77000
472
- },
473
- {
474
- "epoch": 60.94,
475
- "learning_rate": 9.777777777777779e-06,
476
- "loss": 5.0266,
477
- "step": 78000
478
- },
479
- {
480
- "epoch": 61.72,
481
- "learning_rate": 9.333333333333334e-06,
482
- "loss": 5.0235,
483
- "step": 79000
484
- },
485
- {
486
- "epoch": 62.5,
487
- "learning_rate": 8.888888888888888e-06,
488
- "loss": 5.0211,
489
- "step": 80000
490
- },
491
- {
492
- "epoch": 63.28,
493
- "learning_rate": 8.444444444444446e-06,
494
- "loss": 5.0214,
495
- "step": 81000
496
- },
497
- {
498
- "epoch": 64.06,
499
- "learning_rate": 8.000000000000001e-06,
500
- "loss": 5.0176,
501
- "step": 82000
502
- },
503
- {
504
- "epoch": 64.84,
505
- "learning_rate": 7.555555555555556e-06,
506
- "loss": 5.0157,
507
- "step": 83000
508
- },
509
- {
510
- "epoch": 65.62,
511
- "learning_rate": 7.111111111111112e-06,
512
- "loss": 5.014,
513
- "step": 84000
514
- },
515
- {
516
- "epoch": 66.41,
517
- "learning_rate": 6.666666666666667e-06,
518
- "loss": 5.0115,
519
- "step": 85000
520
- },
521
- {
522
- "epoch": 67.19,
523
- "learning_rate": 6.222222222222223e-06,
524
- "loss": 5.0087,
525
- "step": 86000
526
- },
527
- {
528
- "epoch": 67.97,
529
- "learning_rate": 5.777777777777778e-06,
530
- "loss": 5.0105,
531
- "step": 87000
532
- },
533
- {
534
- "epoch": 68.75,
535
- "learning_rate": 5.333333333333334e-06,
536
- "loss": 5.0062,
537
- "step": 88000
538
- },
539
- {
540
- "epoch": 69.53,
541
- "learning_rate": 4.888888888888889e-06,
542
- "loss": 5.0028,
543
- "step": 89000
544
- },
545
- {
546
- "epoch": 70.31,
547
- "learning_rate": 4.444444444444444e-06,
548
- "loss": 5.0066,
549
- "step": 90000
550
- },
551
- {
552
- "epoch": 71.09,
553
- "learning_rate": 4.000000000000001e-06,
554
- "loss": 5.0032,
555
- "step": 91000
556
- },
557
- {
558
- "epoch": 71.88,
559
- "learning_rate": 3.555555555555556e-06,
560
- "loss": 5.0022,
561
- "step": 92000
562
- },
563
- {
564
- "epoch": 72.66,
565
- "learning_rate": 3.1111111111111116e-06,
566
- "loss": 5.0024,
567
- "step": 93000
568
- },
569
- {
570
- "epoch": 73.44,
571
- "learning_rate": 2.666666666666667e-06,
572
- "loss": 5.0007,
573
- "step": 94000
574
- },
575
- {
576
- "epoch": 74.22,
577
- "learning_rate": 2.222222222222222e-06,
578
- "loss": 5.0008,
579
- "step": 95000
580
- },
581
- {
582
- "epoch": 75.0,
583
- "learning_rate": 1.777777777777778e-06,
584
- "loss": 4.9978,
585
- "step": 96000
586
- },
587
- {
588
- "epoch": 75.78,
589
- "learning_rate": 1.3333333333333334e-06,
590
- "loss": 4.9966,
591
- "step": 97000
592
- },
593
- {
594
- "epoch": 76.56,
595
- "learning_rate": 8.88888888888889e-07,
596
- "loss": 4.9994,
597
- "step": 98000
598
- },
599
- {
600
- "epoch": 77.34,
601
- "learning_rate": 4.444444444444445e-07,
602
- "loss": 4.9966,
603
- "step": 99000
604
- },
605
- {
606
- "epoch": 78.12,
607
- "learning_rate": 0.0,
608
- "loss": 4.9976,
609
- "step": 100000
610
- },
611
- {
612
- "epoch": 9.87,
613
- "learning_rate": 2.0842105263157895e-05,
614
- "loss": 5.0155,
615
- "step": 101000
616
- },
617
- {
618
- "epoch": 9.97,
619
- "learning_rate": 2.0631578947368425e-05,
620
- "loss": 5.024,
621
- "step": 102000
622
- },
623
- {
624
- "epoch": 10.06,
625
- "learning_rate": 2.0421052631578952e-05,
626
- "loss": 5.0241,
627
- "step": 103000
628
- },
629
- {
630
- "epoch": 10.16,
631
- "learning_rate": 2.0210526315789475e-05,
632
- "loss": 5.0218,
633
- "step": 104000
634
- },
635
- {
636
- "epoch": 10.26,
637
- "learning_rate": 2e-05,
638
- "loss": 5.0231,
639
- "step": 105000
640
- },
641
- {
642
- "epoch": 10.36,
643
- "learning_rate": 1.9789473684210528e-05,
644
- "loss": 5.0231,
645
- "step": 106000
646
- },
647
- {
648
- "epoch": 10.45,
649
- "learning_rate": 1.9578947368421055e-05,
650
- "loss": 5.0198,
651
- "step": 107000
652
- },
653
- {
654
- "epoch": 10.55,
655
- "learning_rate": 1.936842105263158e-05,
656
- "loss": 5.0158,
657
- "step": 108000
658
- },
659
- {
660
- "epoch": 10.65,
661
- "learning_rate": 1.9157894736842108e-05,
662
- "loss": 5.0122,
663
- "step": 109000
664
- },
665
- {
666
- "epoch": 10.75,
667
- "learning_rate": 1.894736842105263e-05,
668
- "loss": 5.0165,
669
- "step": 110000
670
- },
671
- {
672
- "epoch": 10.85,
673
- "learning_rate": 1.873684210526316e-05,
674
- "loss": 5.0183,
675
- "step": 111000
676
- },
677
- {
678
- "epoch": 10.94,
679
- "learning_rate": 1.8526315789473684e-05,
680
- "loss": 5.01,
681
- "step": 112000
682
- },
683
- {
684
- "epoch": 11.04,
685
- "learning_rate": 1.831578947368421e-05,
686
- "loss": 5.007,
687
- "step": 113000
688
- },
689
- {
690
- "epoch": 11.14,
691
- "learning_rate": 1.810526315789474e-05,
692
- "loss": 5.0059,
693
- "step": 114000
694
- },
695
- {
696
- "epoch": 11.24,
697
- "learning_rate": 1.7894736842105264e-05,
698
- "loss": 5.0089,
699
- "step": 115000
700
- },
701
- {
702
- "epoch": 11.33,
703
- "learning_rate": 1.768421052631579e-05,
704
- "loss": 5.0081,
705
- "step": 116000
706
- },
707
- {
708
- "epoch": 11.43,
709
- "learning_rate": 1.7473684210526317e-05,
710
- "loss": 5.002,
711
- "step": 117000
712
- },
713
- {
714
- "epoch": 11.53,
715
- "learning_rate": 1.7263157894736843e-05,
716
- "loss": 5.0052,
717
- "step": 118000
718
- },
719
- {
720
- "epoch": 11.63,
721
- "learning_rate": 1.705263157894737e-05,
722
- "loss": 5.0079,
723
- "step": 119000
724
- },
725
- {
726
- "epoch": 11.72,
727
- "learning_rate": 1.6842105263157896e-05,
728
- "loss": 4.9985,
729
- "step": 120000
730
- },
731
- {
732
- "epoch": 11.82,
733
- "learning_rate": 1.6631578947368423e-05,
734
- "loss": 5.006,
735
- "step": 121000
736
- },
737
- {
738
- "epoch": 11.92,
739
- "learning_rate": 1.642105263157895e-05,
740
- "loss": 4.9956,
741
- "step": 122000
742
- },
743
- {
744
- "epoch": 12.02,
745
- "learning_rate": 1.6210526315789473e-05,
746
- "loss": 4.9973,
747
- "step": 123000
748
- },
749
- {
750
- "epoch": 12.12,
751
- "learning_rate": 1.6000000000000003e-05,
752
- "loss": 4.9976,
753
- "step": 124000
754
- },
755
- {
756
- "epoch": 12.21,
757
- "learning_rate": 1.578947368421053e-05,
758
- "loss": 4.9882,
759
- "step": 125000
760
- },
761
- {
762
- "epoch": 12.31,
763
- "learning_rate": 1.5578947368421052e-05,
764
- "loss": 4.9908,
765
- "step": 126000
766
- },
767
- {
768
- "epoch": 12.41,
769
- "learning_rate": 1.536842105263158e-05,
770
- "loss": 5.0038,
771
- "step": 127000
772
- },
773
- {
774
- "epoch": 12.51,
775
- "learning_rate": 1.5157894736842107e-05,
776
- "loss": 4.9923,
777
- "step": 128000
778
- },
779
- {
780
- "epoch": 12.6,
781
- "learning_rate": 1.4947368421052632e-05,
782
- "loss": 4.9847,
783
- "step": 129000
784
- },
785
- {
786
- "epoch": 12.7,
787
- "learning_rate": 1.4736842105263159e-05,
788
- "loss": 4.9864,
789
- "step": 130000
790
- },
791
- {
792
- "epoch": 12.8,
793
- "learning_rate": 1.4526315789473687e-05,
794
- "loss": 4.9903,
795
- "step": 131000
796
- },
797
- {
798
- "epoch": 12.9,
799
- "learning_rate": 1.4315789473684212e-05,
800
- "loss": 4.9812,
801
- "step": 132000
802
- },
803
- {
804
- "epoch": 12.99,
805
- "learning_rate": 1.4105263157894738e-05,
806
- "loss": 4.9923,
807
- "step": 133000
808
- },
809
- {
810
- "epoch": 13.09,
811
- "learning_rate": 1.3894736842105265e-05,
812
- "loss": 4.9838,
813
- "step": 134000
814
- },
815
- {
816
- "epoch": 13.19,
817
- "learning_rate": 1.3684210526315791e-05,
818
- "loss": 4.977,
819
- "step": 135000
820
- },
821
- {
822
- "epoch": 13.29,
823
- "learning_rate": 1.3473684210526316e-05,
824
- "loss": 4.9743,
825
- "step": 136000
826
- },
827
- {
828
- "epoch": 13.39,
829
- "learning_rate": 1.3263157894736843e-05,
830
- "loss": 4.9797,
831
- "step": 137000
832
- },
833
- {
834
- "epoch": 13.48,
835
- "learning_rate": 1.305263157894737e-05,
836
- "loss": 4.9915,
837
- "step": 138000
838
- },
839
- {
840
- "epoch": 13.58,
841
- "learning_rate": 1.2842105263157896e-05,
842
- "loss": 4.9762,
843
- "step": 139000
844
- },
845
- {
846
- "epoch": 13.68,
847
- "learning_rate": 1.263157894736842e-05,
848
- "loss": 4.9771,
849
- "step": 140000
850
- },
851
- {
852
- "epoch": 13.78,
853
- "learning_rate": 1.2421052631578949e-05,
854
- "loss": 4.9763,
855
- "step": 141000
856
- },
857
- {
858
- "epoch": 13.87,
859
- "learning_rate": 1.2210526315789475e-05,
860
- "loss": 4.9706,
861
- "step": 142000
862
- },
863
- {
864
- "epoch": 13.97,
865
- "learning_rate": 1.2e-05,
866
- "loss": 4.9731,
867
- "step": 143000
868
- },
869
- {
870
- "epoch": 14.07,
871
- "learning_rate": 1.1789473684210527e-05,
872
- "loss": 4.9736,
873
- "step": 144000
874
- },
875
- {
876
- "epoch": 14.17,
877
- "learning_rate": 1.1578947368421053e-05,
878
- "loss": 4.9704,
879
- "step": 145000
880
- },
881
- {
882
- "epoch": 14.26,
883
- "learning_rate": 1.136842105263158e-05,
884
- "loss": 4.9762,
885
- "step": 146000
886
- },
887
- {
888
- "epoch": 14.36,
889
- "learning_rate": 1.1157894736842105e-05,
890
- "loss": 4.9709,
891
- "step": 147000
892
- },
893
- {
894
- "epoch": 14.46,
895
- "learning_rate": 1.0947368421052633e-05,
896
- "loss": 4.957,
897
- "step": 148000
898
- },
899
- {
900
- "epoch": 14.56,
901
- "learning_rate": 1.073684210526316e-05,
902
- "loss": 4.9666,
903
- "step": 149000
904
- },
905
- {
906
- "epoch": 14.66,
907
- "learning_rate": 1.0526315789473684e-05,
908
- "loss": 4.9623,
909
- "step": 150000
910
- },
911
- {
912
- "epoch": 14.75,
913
- "learning_rate": 1.0315789473684213e-05,
914
- "loss": 4.9613,
915
- "step": 151000
916
- },
917
- {
918
- "epoch": 14.85,
919
- "learning_rate": 1.0105263157894738e-05,
920
- "loss": 4.9642,
921
- "step": 152000
922
- },
923
- {
924
- "epoch": 14.95,
925
- "learning_rate": 9.894736842105264e-06,
926
- "loss": 4.9605,
927
- "step": 153000
928
- },
929
- {
930
- "epoch": 15.05,
931
- "learning_rate": 9.68421052631579e-06,
932
- "loss": 4.9688,
933
- "step": 154000
934
- },
935
- {
936
- "epoch": 15.14,
937
- "learning_rate": 9.473684210526315e-06,
938
- "loss": 4.9635,
939
- "step": 155000
940
- },
941
- {
942
- "epoch": 15.24,
943
- "learning_rate": 9.263157894736842e-06,
944
- "loss": 4.9587,
945
- "step": 156000
946
- },
947
- {
948
- "epoch": 15.34,
949
- "learning_rate": 9.05263157894737e-06,
950
- "loss": 4.9633,
951
- "step": 157000
952
- },
953
- {
954
- "epoch": 15.44,
955
- "learning_rate": 8.842105263157895e-06,
956
- "loss": 4.9613,
957
- "step": 158000
958
- },
959
- {
960
- "epoch": 15.53,
961
- "learning_rate": 8.631578947368422e-06,
962
- "loss": 4.9629,
963
- "step": 159000
964
- },
965
- {
966
- "epoch": 15.63,
967
- "learning_rate": 8.421052631578948e-06,
968
- "loss": 4.9547,
969
- "step": 160000
970
- },
971
- {
972
- "epoch": 15.73,
973
- "learning_rate": 8.210526315789475e-06,
974
- "loss": 4.9541,
975
- "step": 161000
976
- },
977
- {
978
- "epoch": 15.83,
979
- "learning_rate": 8.000000000000001e-06,
980
- "loss": 4.9532,
981
- "step": 162000
982
- },
983
- {
984
- "epoch": 15.93,
985
- "learning_rate": 7.789473684210526e-06,
986
- "loss": 4.9461,
987
- "step": 163000
988
- },
989
- {
990
- "epoch": 16.02,
991
- "learning_rate": 7.578947368421054e-06,
992
- "loss": 4.9518,
993
- "step": 164000
994
- },
995
- {
996
- "epoch": 16.12,
997
- "learning_rate": 7.368421052631579e-06,
998
- "loss": 4.9543,
999
- "step": 165000
1000
- },
1001
- {
1002
- "epoch": 16.22,
1003
- "learning_rate": 7.157894736842106e-06,
1004
- "loss": 4.9526,
1005
- "step": 166000
1006
- },
1007
- {
1008
- "epoch": 16.32,
1009
- "learning_rate": 6.947368421052632e-06,
1010
- "loss": 4.9447,
1011
- "step": 167000
1012
- },
1013
- {
1014
- "epoch": 16.41,
1015
- "learning_rate": 6.736842105263158e-06,
1016
- "loss": 4.9545,
1017
- "step": 168000
1018
- },
1019
- {
1020
- "epoch": 16.51,
1021
- "learning_rate": 6.526315789473685e-06,
1022
- "loss": 4.9448,
1023
- "step": 169000
1024
- },
1025
- {
1026
- "epoch": 16.61,
1027
- "learning_rate": 6.31578947368421e-06,
1028
- "loss": 4.9521,
1029
- "step": 170000
1030
- },
1031
- {
1032
- "epoch": 16.71,
1033
- "learning_rate": 6.105263157894738e-06,
1034
- "loss": 4.9503,
1035
- "step": 171000
1036
- },
1037
- {
1038
- "epoch": 16.81,
1039
- "learning_rate": 5.8947368421052634e-06,
1040
- "loss": 4.9537,
1041
- "step": 172000
1042
- },
1043
- {
1044
- "epoch": 16.9,
1045
- "learning_rate": 5.68421052631579e-06,
1046
- "loss": 4.9446,
1047
- "step": 173000
1048
- },
1049
- {
1050
- "epoch": 17.0,
1051
- "learning_rate": 5.4736842105263165e-06,
1052
- "loss": 4.9397,
1053
- "step": 174000
1054
- },
1055
- {
1056
- "epoch": 17.1,
1057
- "learning_rate": 5.263157894736842e-06,
1058
- "loss": 4.9541,
1059
- "step": 175000
1060
- },
1061
- {
1062
- "epoch": 17.2,
1063
- "learning_rate": 5.052631578947369e-06,
1064
- "loss": 4.9449,
1065
- "step": 176000
1066
- },
1067
- {
1068
- "epoch": 17.29,
1069
- "learning_rate": 4.842105263157895e-06,
1070
- "loss": 4.9447,
1071
- "step": 177000
1072
- },
1073
- {
1074
- "epoch": 17.39,
1075
- "learning_rate": 4.631578947368421e-06,
1076
- "loss": 4.9344,
1077
- "step": 178000
1078
- },
1079
- {
1080
- "epoch": 17.49,
1081
- "learning_rate": 4.4210526315789476e-06,
1082
- "loss": 4.933,
1083
- "step": 179000
1084
- },
1085
- {
1086
- "epoch": 17.59,
1087
- "learning_rate": 4.210526315789474e-06,
1088
- "loss": 4.9397,
1089
- "step": 180000
1090
- },
1091
- {
1092
- "epoch": 17.68,
1093
- "learning_rate": 4.000000000000001e-06,
1094
- "loss": 4.9431,
1095
- "step": 181000
1096
- },
1097
- {
1098
- "epoch": 17.78,
1099
- "learning_rate": 3.789473684210527e-06,
1100
- "loss": 4.9378,
1101
- "step": 182000
1102
- },
1103
- {
1104
- "epoch": 17.88,
1105
- "learning_rate": 3.578947368421053e-06,
1106
- "loss": 4.9399,
1107
- "step": 183000
1108
- },
1109
- {
1110
- "epoch": 17.98,
1111
- "learning_rate": 3.368421052631579e-06,
1112
- "loss": 4.9481,
1113
- "step": 184000
1114
- },
1115
- {
1116
- "epoch": 18.08,
1117
- "learning_rate": 3.157894736842105e-06,
1118
- "loss": 4.9399,
1119
- "step": 185000
1120
- },
1121
- {
1122
- "epoch": 18.17,
1123
- "learning_rate": 2.9473684210526317e-06,
1124
- "loss": 4.9365,
1125
- "step": 186000
1126
- },
1127
- {
1128
- "epoch": 18.27,
1129
- "learning_rate": 2.7368421052631583e-06,
1130
- "loss": 4.9307,
1131
- "step": 187000
1132
- },
1133
- {
1134
- "epoch": 18.37,
1135
- "learning_rate": 2.5263157894736844e-06,
1136
- "loss": 4.9391,
1137
- "step": 188000
1138
- },
1139
- {
1140
- "epoch": 18.47,
1141
- "learning_rate": 2.3157894736842105e-06,
1142
- "loss": 4.9331,
1143
- "step": 189000
1144
- },
1145
- {
1146
- "epoch": 18.56,
1147
- "learning_rate": 2.105263157894737e-06,
1148
- "loss": 4.948,
1149
- "step": 190000
1150
- },
1151
- {
1152
- "epoch": 18.66,
1153
- "learning_rate": 1.8947368421052634e-06,
1154
- "loss": 4.9407,
1155
- "step": 191000
1156
- },
1157
- {
1158
- "epoch": 18.76,
1159
- "learning_rate": 1.6842105263157895e-06,
1160
- "loss": 4.944,
1161
- "step": 192000
1162
- }
1163
- ],
1164
- "logging_steps": 1000,
1165
- "max_steps": 200000,
1166
- "num_train_epochs": 20,
1167
- "save_steps": 1000,
1168
- "total_flos": 2.2531937955260137e+19,
1169
- "trial_name": null,
1170
- "trial_params": null
1171
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/checkpoint-192000/training_args.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:c9ff5d6388a59708cb6305a7fab984d438e385b1835680f9629b90c23c5cf704
3
- size 5137
 
 
 
 
len2_5120/checkpoint-193000/config.json DELETED
@@ -1,25 +0,0 @@
1
- {
2
- "architectures": [
3
- "BertForMaskedLM"
4
- ],
5
- "attention_probs_dropout_prob": 0.1,
6
- "classifier_dropout": null,
7
- "hidden_act": "gelu",
8
- "hidden_dropout_prob": 0.1,
9
- "hidden_size": 768,
10
- "initializer_range": 0.02,
11
- "intermediate_size": 3072,
12
- "layer_norm_eps": 1e-12,
13
- "max_length": 512,
14
- "max_position_embeddings": 512,
15
- "model_type": "bert",
16
- "num_attention_heads": 12,
17
- "num_hidden_layers": 12,
18
- "pad_token_id": 0,
19
- "position_embedding_type": "absolute",
20
- "torch_dtype": "float32",
21
- "transformers_version": "4.35.2",
22
- "type_vocab_size": 2,
23
- "use_cache": true,
24
- "vocab_size": 5120
25
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/checkpoint-193000/generation_config.json DELETED
@@ -1,6 +0,0 @@
1
- {
2
- "_from_model_config": true,
3
- "max_length": 512,
4
- "pad_token_id": 0,
5
- "transformers_version": "4.35.2"
6
- }
 
 
 
 
 
 
 
len2_5120/checkpoint-193000/model.safetensors DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:2967019014b91b75c93d4a3368eade80f165f36a45202547cb4fffa4d2bd3dcc
3
- size 359944256
 
 
 
 
len2_5120/checkpoint-193000/optimizer.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:fe6d3dd22b82273c51f5e1ef00d4b8b80373bff065365e28532536ac9459fe50
3
- size 720010699
 
 
 
 
len2_5120/checkpoint-193000/rng_state.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:a14699627c750f517769353fa39bf167eec026587c76244f12abfd0f16cabde3
3
- size 14645
 
 
 
 
len2_5120/checkpoint-193000/scheduler.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:193ff94d9f24279c5f8926a3af20ad4747ac16bb078d5e13b935e0b3be972c5b
3
- size 1465
 
 
 
 
len2_5120/checkpoint-193000/trainer_state.json DELETED
@@ -1,1177 +0,0 @@
1
- {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 18.85686370297997,
5
- "eval_steps": 500,
6
- "global_step": 193000,
7
- "is_hyper_param_search": false,
8
- "is_local_process_zero": true,
9
- "is_world_process_zero": true,
10
- "log_history": [
11
- {
12
- "epoch": 0.78,
13
- "learning_rate": 4.000000000000001e-06,
14
- "loss": 7.5832,
15
- "step": 1000
16
- },
17
- {
18
- "epoch": 1.56,
19
- "learning_rate": 8.000000000000001e-06,
20
- "loss": 6.9724,
21
- "step": 2000
22
- },
23
- {
24
- "epoch": 2.34,
25
- "learning_rate": 1.2e-05,
26
- "loss": 6.7695,
27
- "step": 3000
28
- },
29
- {
30
- "epoch": 3.12,
31
- "learning_rate": 1.6000000000000003e-05,
32
- "loss": 6.7002,
33
- "step": 4000
34
- },
35
- {
36
- "epoch": 3.91,
37
- "learning_rate": 2e-05,
38
- "loss": 6.666,
39
- "step": 5000
40
- },
41
- {
42
- "epoch": 4.69,
43
- "learning_rate": 2.4e-05,
44
- "loss": 6.6503,
45
- "step": 6000
46
- },
47
- {
48
- "epoch": 5.47,
49
- "learning_rate": 2.8e-05,
50
- "loss": 6.6407,
51
- "step": 7000
52
- },
53
- {
54
- "epoch": 6.25,
55
- "learning_rate": 3.2000000000000005e-05,
56
- "loss": 6.6277,
57
- "step": 8000
58
- },
59
- {
60
- "epoch": 7.03,
61
- "learning_rate": 3.6e-05,
62
- "loss": 6.6209,
63
- "step": 9000
64
- },
65
- {
66
- "epoch": 7.81,
67
- "learning_rate": 4e-05,
68
- "loss": 6.6135,
69
- "step": 10000
70
- },
71
- {
72
- "epoch": 8.59,
73
- "learning_rate": 3.955555555555556e-05,
74
- "loss": 6.6084,
75
- "step": 11000
76
- },
77
- {
78
- "epoch": 9.38,
79
- "learning_rate": 3.9111111111111115e-05,
80
- "loss": 6.6041,
81
- "step": 12000
82
- },
83
- {
84
- "epoch": 10.16,
85
- "learning_rate": 3.866666666666667e-05,
86
- "loss": 6.6013,
87
- "step": 13000
88
- },
89
- {
90
- "epoch": 10.94,
91
- "learning_rate": 3.8222222222222226e-05,
92
- "loss": 6.5994,
93
- "step": 14000
94
- },
95
- {
96
- "epoch": 11.72,
97
- "learning_rate": 3.777777777777778e-05,
98
- "loss": 6.594,
99
- "step": 15000
100
- },
101
- {
102
- "epoch": 12.5,
103
- "learning_rate": 3.733333333333334e-05,
104
- "loss": 6.5925,
105
- "step": 16000
106
- },
107
- {
108
- "epoch": 13.28,
109
- "learning_rate": 3.6888888888888896e-05,
110
- "loss": 6.5893,
111
- "step": 17000
112
- },
113
- {
114
- "epoch": 14.06,
115
- "learning_rate": 3.644444444444445e-05,
116
- "loss": 6.5875,
117
- "step": 18000
118
- },
119
- {
120
- "epoch": 14.84,
121
- "learning_rate": 3.6e-05,
122
- "loss": 6.5232,
123
- "step": 19000
124
- },
125
- {
126
- "epoch": 15.62,
127
- "learning_rate": 3.555555555555555e-05,
128
- "loss": 6.4358,
129
- "step": 20000
130
- },
131
- {
132
- "epoch": 16.41,
133
- "learning_rate": 3.511111111111111e-05,
134
- "loss": 6.3712,
135
- "step": 21000
136
- },
137
- {
138
- "epoch": 17.19,
139
- "learning_rate": 3.466666666666667e-05,
140
- "loss": 6.311,
141
- "step": 22000
142
- },
143
- {
144
- "epoch": 17.97,
145
- "learning_rate": 3.4222222222222224e-05,
146
- "loss": 6.2493,
147
- "step": 23000
148
- },
149
- {
150
- "epoch": 18.75,
151
- "learning_rate": 3.377777777777778e-05,
152
- "loss": 6.1597,
153
- "step": 24000
154
- },
155
- {
156
- "epoch": 19.53,
157
- "learning_rate": 3.3333333333333335e-05,
158
- "loss": 6.0276,
159
- "step": 25000
160
- },
161
- {
162
- "epoch": 20.31,
163
- "learning_rate": 3.288888888888889e-05,
164
- "loss": 5.8317,
165
- "step": 26000
166
- },
167
- {
168
- "epoch": 21.09,
169
- "learning_rate": 3.2444444444444446e-05,
170
- "loss": 5.6782,
171
- "step": 27000
172
- },
173
- {
174
- "epoch": 21.88,
175
- "learning_rate": 3.2000000000000005e-05,
176
- "loss": 5.5975,
177
- "step": 28000
178
- },
179
- {
180
- "epoch": 22.66,
181
- "learning_rate": 3.155555555555556e-05,
182
- "loss": 5.5369,
183
- "step": 29000
184
- },
185
- {
186
- "epoch": 23.44,
187
- "learning_rate": 3.111111111111112e-05,
188
- "loss": 5.4921,
189
- "step": 30000
190
- },
191
- {
192
- "epoch": 24.22,
193
- "learning_rate": 3.066666666666667e-05,
194
- "loss": 5.4554,
195
- "step": 31000
196
- },
197
- {
198
- "epoch": 25.0,
199
- "learning_rate": 3.0222222222222225e-05,
200
- "loss": 5.4237,
201
- "step": 32000
202
- },
203
- {
204
- "epoch": 25.78,
205
- "learning_rate": 2.977777777777778e-05,
206
- "loss": 5.3938,
207
- "step": 33000
208
- },
209
- {
210
- "epoch": 26.56,
211
- "learning_rate": 2.9333333333333333e-05,
212
- "loss": 5.3701,
213
- "step": 34000
214
- },
215
- {
216
- "epoch": 27.34,
217
- "learning_rate": 2.888888888888889e-05,
218
- "loss": 5.3452,
219
- "step": 35000
220
- },
221
- {
222
- "epoch": 28.12,
223
- "learning_rate": 2.8444444444444447e-05,
224
- "loss": 5.327,
225
- "step": 36000
226
- },
227
- {
228
- "epoch": 28.91,
229
- "learning_rate": 2.8e-05,
230
- "loss": 5.3063,
231
- "step": 37000
232
- },
233
- {
234
- "epoch": 29.69,
235
- "learning_rate": 2.755555555555556e-05,
236
- "loss": 5.2905,
237
- "step": 38000
238
- },
239
- {
240
- "epoch": 30.47,
241
- "learning_rate": 2.7111111111111114e-05,
242
- "loss": 5.2747,
243
- "step": 39000
244
- },
245
- {
246
- "epoch": 31.25,
247
- "learning_rate": 2.6666666666666667e-05,
248
- "loss": 5.2596,
249
- "step": 40000
250
- },
251
- {
252
- "epoch": 32.03,
253
- "learning_rate": 2.6222222222222226e-05,
254
- "loss": 5.2483,
255
- "step": 41000
256
- },
257
- {
258
- "epoch": 32.81,
259
- "learning_rate": 2.577777777777778e-05,
260
- "loss": 5.2336,
261
- "step": 42000
262
- },
263
- {
264
- "epoch": 33.59,
265
- "learning_rate": 2.5333333333333334e-05,
266
- "loss": 5.2214,
267
- "step": 43000
268
- },
269
- {
270
- "epoch": 34.38,
271
- "learning_rate": 2.4888888888888893e-05,
272
- "loss": 5.2103,
273
- "step": 44000
274
- },
275
- {
276
- "epoch": 35.16,
277
- "learning_rate": 2.444444444444445e-05,
278
- "loss": 5.2015,
279
- "step": 45000
280
- },
281
- {
282
- "epoch": 35.94,
283
- "learning_rate": 2.4e-05,
284
- "loss": 5.1907,
285
- "step": 46000
286
- },
287
- {
288
- "epoch": 36.72,
289
- "learning_rate": 2.355555555555556e-05,
290
- "loss": 5.1801,
291
- "step": 47000
292
- },
293
- {
294
- "epoch": 37.5,
295
- "learning_rate": 2.3111111111111112e-05,
296
- "loss": 5.1713,
297
- "step": 48000
298
- },
299
- {
300
- "epoch": 38.28,
301
- "learning_rate": 2.2666666666666668e-05,
302
- "loss": 5.1645,
303
- "step": 49000
304
- },
305
- {
306
- "epoch": 39.06,
307
- "learning_rate": 2.2222222222222227e-05,
308
- "loss": 5.1548,
309
- "step": 50000
310
- },
311
- {
312
- "epoch": 39.84,
313
- "learning_rate": 2.177777777777778e-05,
314
- "loss": 5.1462,
315
- "step": 51000
316
- },
317
- {
318
- "epoch": 40.62,
319
- "learning_rate": 2.1333333333333335e-05,
320
- "loss": 5.1417,
321
- "step": 52000
322
- },
323
- {
324
- "epoch": 41.41,
325
- "learning_rate": 2.088888888888889e-05,
326
- "loss": 5.1305,
327
- "step": 53000
328
- },
329
- {
330
- "epoch": 42.19,
331
- "learning_rate": 2.0444444444444446e-05,
332
- "loss": 5.1287,
333
- "step": 54000
334
- },
335
- {
336
- "epoch": 42.97,
337
- "learning_rate": 2e-05,
338
- "loss": 5.1202,
339
- "step": 55000
340
- },
341
- {
342
- "epoch": 43.75,
343
- "learning_rate": 1.9555555555555557e-05,
344
- "loss": 5.1139,
345
- "step": 56000
346
- },
347
- {
348
- "epoch": 44.53,
349
- "learning_rate": 1.9111111111111113e-05,
350
- "loss": 5.1081,
351
- "step": 57000
352
- },
353
- {
354
- "epoch": 45.31,
355
- "learning_rate": 1.866666666666667e-05,
356
- "loss": 5.1032,
357
- "step": 58000
358
- },
359
- {
360
- "epoch": 46.09,
361
- "learning_rate": 1.8222222222222224e-05,
362
- "loss": 5.0968,
363
- "step": 59000
364
- },
365
- {
366
- "epoch": 46.88,
367
- "learning_rate": 1.7777777777777777e-05,
368
- "loss": 5.0914,
369
- "step": 60000
370
- },
371
- {
372
- "epoch": 47.66,
373
- "learning_rate": 1.7333333333333336e-05,
374
- "loss": 5.085,
375
- "step": 61000
376
- },
377
- {
378
- "epoch": 48.44,
379
- "learning_rate": 1.688888888888889e-05,
380
- "loss": 5.0806,
381
- "step": 62000
382
- },
383
- {
384
- "epoch": 49.22,
385
- "learning_rate": 1.6444444444444444e-05,
386
- "loss": 5.0769,
387
- "step": 63000
388
- },
389
- {
390
- "epoch": 50.0,
391
- "learning_rate": 1.6000000000000003e-05,
392
- "loss": 5.0728,
393
- "step": 64000
394
- },
395
- {
396
- "epoch": 50.78,
397
- "learning_rate": 1.555555555555556e-05,
398
- "loss": 5.0692,
399
- "step": 65000
400
- },
401
- {
402
- "epoch": 51.56,
403
- "learning_rate": 1.5111111111111112e-05,
404
- "loss": 5.065,
405
- "step": 66000
406
- },
407
- {
408
- "epoch": 52.34,
409
- "learning_rate": 1.4666666666666666e-05,
410
- "loss": 5.0594,
411
- "step": 67000
412
- },
413
- {
414
- "epoch": 53.12,
415
- "learning_rate": 1.4222222222222224e-05,
416
- "loss": 5.0549,
417
- "step": 68000
418
- },
419
- {
420
- "epoch": 53.91,
421
- "learning_rate": 1.377777777777778e-05,
422
- "loss": 5.0538,
423
- "step": 69000
424
- },
425
- {
426
- "epoch": 54.69,
427
- "learning_rate": 1.3333333333333333e-05,
428
- "loss": 5.0482,
429
- "step": 70000
430
- },
431
- {
432
- "epoch": 55.47,
433
- "learning_rate": 1.288888888888889e-05,
434
- "loss": 5.0475,
435
- "step": 71000
436
- },
437
- {
438
- "epoch": 56.25,
439
- "learning_rate": 1.2444444444444446e-05,
440
- "loss": 5.0399,
441
- "step": 72000
442
- },
443
- {
444
- "epoch": 57.03,
445
- "learning_rate": 1.2e-05,
446
- "loss": 5.0406,
447
- "step": 73000
448
- },
449
- {
450
- "epoch": 57.81,
451
- "learning_rate": 1.1555555555555556e-05,
452
- "loss": 5.037,
453
- "step": 74000
454
- },
455
- {
456
- "epoch": 58.59,
457
- "learning_rate": 1.1111111111111113e-05,
458
- "loss": 5.0326,
459
- "step": 75000
460
- },
461
- {
462
- "epoch": 59.38,
463
- "learning_rate": 1.0666666666666667e-05,
464
- "loss": 5.031,
465
- "step": 76000
466
- },
467
- {
468
- "epoch": 60.16,
469
- "learning_rate": 1.0222222222222223e-05,
470
- "loss": 5.0291,
471
- "step": 77000
472
- },
473
- {
474
- "epoch": 60.94,
475
- "learning_rate": 9.777777777777779e-06,
476
- "loss": 5.0266,
477
- "step": 78000
478
- },
479
- {
480
- "epoch": 61.72,
481
- "learning_rate": 9.333333333333334e-06,
482
- "loss": 5.0235,
483
- "step": 79000
484
- },
485
- {
486
- "epoch": 62.5,
487
- "learning_rate": 8.888888888888888e-06,
488
- "loss": 5.0211,
489
- "step": 80000
490
- },
491
- {
492
- "epoch": 63.28,
493
- "learning_rate": 8.444444444444446e-06,
494
- "loss": 5.0214,
495
- "step": 81000
496
- },
497
- {
498
- "epoch": 64.06,
499
- "learning_rate": 8.000000000000001e-06,
500
- "loss": 5.0176,
501
- "step": 82000
502
- },
503
- {
504
- "epoch": 64.84,
505
- "learning_rate": 7.555555555555556e-06,
506
- "loss": 5.0157,
507
- "step": 83000
508
- },
509
- {
510
- "epoch": 65.62,
511
- "learning_rate": 7.111111111111112e-06,
512
- "loss": 5.014,
513
- "step": 84000
514
- },
515
- {
516
- "epoch": 66.41,
517
- "learning_rate": 6.666666666666667e-06,
518
- "loss": 5.0115,
519
- "step": 85000
520
- },
521
- {
522
- "epoch": 67.19,
523
- "learning_rate": 6.222222222222223e-06,
524
- "loss": 5.0087,
525
- "step": 86000
526
- },
527
- {
528
- "epoch": 67.97,
529
- "learning_rate": 5.777777777777778e-06,
530
- "loss": 5.0105,
531
- "step": 87000
532
- },
533
- {
534
- "epoch": 68.75,
535
- "learning_rate": 5.333333333333334e-06,
536
- "loss": 5.0062,
537
- "step": 88000
538
- },
539
- {
540
- "epoch": 69.53,
541
- "learning_rate": 4.888888888888889e-06,
542
- "loss": 5.0028,
543
- "step": 89000
544
- },
545
- {
546
- "epoch": 70.31,
547
- "learning_rate": 4.444444444444444e-06,
548
- "loss": 5.0066,
549
- "step": 90000
550
- },
551
- {
552
- "epoch": 71.09,
553
- "learning_rate": 4.000000000000001e-06,
554
- "loss": 5.0032,
555
- "step": 91000
556
- },
557
- {
558
- "epoch": 71.88,
559
- "learning_rate": 3.555555555555556e-06,
560
- "loss": 5.0022,
561
- "step": 92000
562
- },
563
- {
564
- "epoch": 72.66,
565
- "learning_rate": 3.1111111111111116e-06,
566
- "loss": 5.0024,
567
- "step": 93000
568
- },
569
- {
570
- "epoch": 73.44,
571
- "learning_rate": 2.666666666666667e-06,
572
- "loss": 5.0007,
573
- "step": 94000
574
- },
575
- {
576
- "epoch": 74.22,
577
- "learning_rate": 2.222222222222222e-06,
578
- "loss": 5.0008,
579
- "step": 95000
580
- },
581
- {
582
- "epoch": 75.0,
583
- "learning_rate": 1.777777777777778e-06,
584
- "loss": 4.9978,
585
- "step": 96000
586
- },
587
- {
588
- "epoch": 75.78,
589
- "learning_rate": 1.3333333333333334e-06,
590
- "loss": 4.9966,
591
- "step": 97000
592
- },
593
- {
594
- "epoch": 76.56,
595
- "learning_rate": 8.88888888888889e-07,
596
- "loss": 4.9994,
597
- "step": 98000
598
- },
599
- {
600
- "epoch": 77.34,
601
- "learning_rate": 4.444444444444445e-07,
602
- "loss": 4.9966,
603
- "step": 99000
604
- },
605
- {
606
- "epoch": 78.12,
607
- "learning_rate": 0.0,
608
- "loss": 4.9976,
609
- "step": 100000
610
- },
611
- {
612
- "epoch": 9.87,
613
- "learning_rate": 2.0842105263157895e-05,
614
- "loss": 5.0155,
615
- "step": 101000
616
- },
617
- {
618
- "epoch": 9.97,
619
- "learning_rate": 2.0631578947368425e-05,
620
- "loss": 5.024,
621
- "step": 102000
622
- },
623
- {
624
- "epoch": 10.06,
625
- "learning_rate": 2.0421052631578952e-05,
626
- "loss": 5.0241,
627
- "step": 103000
628
- },
629
- {
630
- "epoch": 10.16,
631
- "learning_rate": 2.0210526315789475e-05,
632
- "loss": 5.0218,
633
- "step": 104000
634
- },
635
- {
636
- "epoch": 10.26,
637
- "learning_rate": 2e-05,
638
- "loss": 5.0231,
639
- "step": 105000
640
- },
641
- {
642
- "epoch": 10.36,
643
- "learning_rate": 1.9789473684210528e-05,
644
- "loss": 5.0231,
645
- "step": 106000
646
- },
647
- {
648
- "epoch": 10.45,
649
- "learning_rate": 1.9578947368421055e-05,
650
- "loss": 5.0198,
651
- "step": 107000
652
- },
653
- {
654
- "epoch": 10.55,
655
- "learning_rate": 1.936842105263158e-05,
656
- "loss": 5.0158,
657
- "step": 108000
658
- },
659
- {
660
- "epoch": 10.65,
661
- "learning_rate": 1.9157894736842108e-05,
662
- "loss": 5.0122,
663
- "step": 109000
664
- },
665
- {
666
- "epoch": 10.75,
667
- "learning_rate": 1.894736842105263e-05,
668
- "loss": 5.0165,
669
- "step": 110000
670
- },
671
- {
672
- "epoch": 10.85,
673
- "learning_rate": 1.873684210526316e-05,
674
- "loss": 5.0183,
675
- "step": 111000
676
- },
677
- {
678
- "epoch": 10.94,
679
- "learning_rate": 1.8526315789473684e-05,
680
- "loss": 5.01,
681
- "step": 112000
682
- },
683
- {
684
- "epoch": 11.04,
685
- "learning_rate": 1.831578947368421e-05,
686
- "loss": 5.007,
687
- "step": 113000
688
- },
689
- {
690
- "epoch": 11.14,
691
- "learning_rate": 1.810526315789474e-05,
692
- "loss": 5.0059,
693
- "step": 114000
694
- },
695
- {
696
- "epoch": 11.24,
697
- "learning_rate": 1.7894736842105264e-05,
698
- "loss": 5.0089,
699
- "step": 115000
700
- },
701
- {
702
- "epoch": 11.33,
703
- "learning_rate": 1.768421052631579e-05,
704
- "loss": 5.0081,
705
- "step": 116000
706
- },
707
- {
708
- "epoch": 11.43,
709
- "learning_rate": 1.7473684210526317e-05,
710
- "loss": 5.002,
711
- "step": 117000
712
- },
713
- {
714
- "epoch": 11.53,
715
- "learning_rate": 1.7263157894736843e-05,
716
- "loss": 5.0052,
717
- "step": 118000
718
- },
719
- {
720
- "epoch": 11.63,
721
- "learning_rate": 1.705263157894737e-05,
722
- "loss": 5.0079,
723
- "step": 119000
724
- },
725
- {
726
- "epoch": 11.72,
727
- "learning_rate": 1.6842105263157896e-05,
728
- "loss": 4.9985,
729
- "step": 120000
730
- },
731
- {
732
- "epoch": 11.82,
733
- "learning_rate": 1.6631578947368423e-05,
734
- "loss": 5.006,
735
- "step": 121000
736
- },
737
- {
738
- "epoch": 11.92,
739
- "learning_rate": 1.642105263157895e-05,
740
- "loss": 4.9956,
741
- "step": 122000
742
- },
743
- {
744
- "epoch": 12.02,
745
- "learning_rate": 1.6210526315789473e-05,
746
- "loss": 4.9973,
747
- "step": 123000
748
- },
749
- {
750
- "epoch": 12.12,
751
- "learning_rate": 1.6000000000000003e-05,
752
- "loss": 4.9976,
753
- "step": 124000
754
- },
755
- {
756
- "epoch": 12.21,
757
- "learning_rate": 1.578947368421053e-05,
758
- "loss": 4.9882,
759
- "step": 125000
760
- },
761
- {
762
- "epoch": 12.31,
763
- "learning_rate": 1.5578947368421052e-05,
764
- "loss": 4.9908,
765
- "step": 126000
766
- },
767
- {
768
- "epoch": 12.41,
769
- "learning_rate": 1.536842105263158e-05,
770
- "loss": 5.0038,
771
- "step": 127000
772
- },
773
- {
774
- "epoch": 12.51,
775
- "learning_rate": 1.5157894736842107e-05,
776
- "loss": 4.9923,
777
- "step": 128000
778
- },
779
- {
780
- "epoch": 12.6,
781
- "learning_rate": 1.4947368421052632e-05,
782
- "loss": 4.9847,
783
- "step": 129000
784
- },
785
- {
786
- "epoch": 12.7,
787
- "learning_rate": 1.4736842105263159e-05,
788
- "loss": 4.9864,
789
- "step": 130000
790
- },
791
- {
792
- "epoch": 12.8,
793
- "learning_rate": 1.4526315789473687e-05,
794
- "loss": 4.9903,
795
- "step": 131000
796
- },
797
- {
798
- "epoch": 12.9,
799
- "learning_rate": 1.4315789473684212e-05,
800
- "loss": 4.9812,
801
- "step": 132000
802
- },
803
- {
804
- "epoch": 12.99,
805
- "learning_rate": 1.4105263157894738e-05,
806
- "loss": 4.9923,
807
- "step": 133000
808
- },
809
- {
810
- "epoch": 13.09,
811
- "learning_rate": 1.3894736842105265e-05,
812
- "loss": 4.9838,
813
- "step": 134000
814
- },
815
- {
816
- "epoch": 13.19,
817
- "learning_rate": 1.3684210526315791e-05,
818
- "loss": 4.977,
819
- "step": 135000
820
- },
821
- {
822
- "epoch": 13.29,
823
- "learning_rate": 1.3473684210526316e-05,
824
- "loss": 4.9743,
825
- "step": 136000
826
- },
827
- {
828
- "epoch": 13.39,
829
- "learning_rate": 1.3263157894736843e-05,
830
- "loss": 4.9797,
831
- "step": 137000
832
- },
833
- {
834
- "epoch": 13.48,
835
- "learning_rate": 1.305263157894737e-05,
836
- "loss": 4.9915,
837
- "step": 138000
838
- },
839
- {
840
- "epoch": 13.58,
841
- "learning_rate": 1.2842105263157896e-05,
842
- "loss": 4.9762,
843
- "step": 139000
844
- },
845
- {
846
- "epoch": 13.68,
847
- "learning_rate": 1.263157894736842e-05,
848
- "loss": 4.9771,
849
- "step": 140000
850
- },
851
- {
852
- "epoch": 13.78,
853
- "learning_rate": 1.2421052631578949e-05,
854
- "loss": 4.9763,
855
- "step": 141000
856
- },
857
- {
858
- "epoch": 13.87,
859
- "learning_rate": 1.2210526315789475e-05,
860
- "loss": 4.9706,
861
- "step": 142000
862
- },
863
- {
864
- "epoch": 13.97,
865
- "learning_rate": 1.2e-05,
866
- "loss": 4.9731,
867
- "step": 143000
868
- },
869
- {
870
- "epoch": 14.07,
871
- "learning_rate": 1.1789473684210527e-05,
872
- "loss": 4.9736,
873
- "step": 144000
874
- },
875
- {
876
- "epoch": 14.17,
877
- "learning_rate": 1.1578947368421053e-05,
878
- "loss": 4.9704,
879
- "step": 145000
880
- },
881
- {
882
- "epoch": 14.26,
883
- "learning_rate": 1.136842105263158e-05,
884
- "loss": 4.9762,
885
- "step": 146000
886
- },
887
- {
888
- "epoch": 14.36,
889
- "learning_rate": 1.1157894736842105e-05,
890
- "loss": 4.9709,
891
- "step": 147000
892
- },
893
- {
894
- "epoch": 14.46,
895
- "learning_rate": 1.0947368421052633e-05,
896
- "loss": 4.957,
897
- "step": 148000
898
- },
899
- {
900
- "epoch": 14.56,
901
- "learning_rate": 1.073684210526316e-05,
902
- "loss": 4.9666,
903
- "step": 149000
904
- },
905
- {
906
- "epoch": 14.66,
907
- "learning_rate": 1.0526315789473684e-05,
908
- "loss": 4.9623,
909
- "step": 150000
910
- },
911
- {
912
- "epoch": 14.75,
913
- "learning_rate": 1.0315789473684213e-05,
914
- "loss": 4.9613,
915
- "step": 151000
916
- },
917
- {
918
- "epoch": 14.85,
919
- "learning_rate": 1.0105263157894738e-05,
920
- "loss": 4.9642,
921
- "step": 152000
922
- },
923
- {
924
- "epoch": 14.95,
925
- "learning_rate": 9.894736842105264e-06,
926
- "loss": 4.9605,
927
- "step": 153000
928
- },
929
- {
930
- "epoch": 15.05,
931
- "learning_rate": 9.68421052631579e-06,
932
- "loss": 4.9688,
933
- "step": 154000
934
- },
935
- {
936
- "epoch": 15.14,
937
- "learning_rate": 9.473684210526315e-06,
938
- "loss": 4.9635,
939
- "step": 155000
940
- },
941
- {
942
- "epoch": 15.24,
943
- "learning_rate": 9.263157894736842e-06,
944
- "loss": 4.9587,
945
- "step": 156000
946
- },
947
- {
948
- "epoch": 15.34,
949
- "learning_rate": 9.05263157894737e-06,
950
- "loss": 4.9633,
951
- "step": 157000
952
- },
953
- {
954
- "epoch": 15.44,
955
- "learning_rate": 8.842105263157895e-06,
956
- "loss": 4.9613,
957
- "step": 158000
958
- },
959
- {
960
- "epoch": 15.53,
961
- "learning_rate": 8.631578947368422e-06,
962
- "loss": 4.9629,
963
- "step": 159000
964
- },
965
- {
966
- "epoch": 15.63,
967
- "learning_rate": 8.421052631578948e-06,
968
- "loss": 4.9547,
969
- "step": 160000
970
- },
971
- {
972
- "epoch": 15.73,
973
- "learning_rate": 8.210526315789475e-06,
974
- "loss": 4.9541,
975
- "step": 161000
976
- },
977
- {
978
- "epoch": 15.83,
979
- "learning_rate": 8.000000000000001e-06,
980
- "loss": 4.9532,
981
- "step": 162000
982
- },
983
- {
984
- "epoch": 15.93,
985
- "learning_rate": 7.789473684210526e-06,
986
- "loss": 4.9461,
987
- "step": 163000
988
- },
989
- {
990
- "epoch": 16.02,
991
- "learning_rate": 7.578947368421054e-06,
992
- "loss": 4.9518,
993
- "step": 164000
994
- },
995
- {
996
- "epoch": 16.12,
997
- "learning_rate": 7.368421052631579e-06,
998
- "loss": 4.9543,
999
- "step": 165000
1000
- },
1001
- {
1002
- "epoch": 16.22,
1003
- "learning_rate": 7.157894736842106e-06,
1004
- "loss": 4.9526,
1005
- "step": 166000
1006
- },
1007
- {
1008
- "epoch": 16.32,
1009
- "learning_rate": 6.947368421052632e-06,
1010
- "loss": 4.9447,
1011
- "step": 167000
1012
- },
1013
- {
1014
- "epoch": 16.41,
1015
- "learning_rate": 6.736842105263158e-06,
1016
- "loss": 4.9545,
1017
- "step": 168000
1018
- },
1019
- {
1020
- "epoch": 16.51,
1021
- "learning_rate": 6.526315789473685e-06,
1022
- "loss": 4.9448,
1023
- "step": 169000
1024
- },
1025
- {
1026
- "epoch": 16.61,
1027
- "learning_rate": 6.31578947368421e-06,
1028
- "loss": 4.9521,
1029
- "step": 170000
1030
- },
1031
- {
1032
- "epoch": 16.71,
1033
- "learning_rate": 6.105263157894738e-06,
1034
- "loss": 4.9503,
1035
- "step": 171000
1036
- },
1037
- {
1038
- "epoch": 16.81,
1039
- "learning_rate": 5.8947368421052634e-06,
1040
- "loss": 4.9537,
1041
- "step": 172000
1042
- },
1043
- {
1044
- "epoch": 16.9,
1045
- "learning_rate": 5.68421052631579e-06,
1046
- "loss": 4.9446,
1047
- "step": 173000
1048
- },
1049
- {
1050
- "epoch": 17.0,
1051
- "learning_rate": 5.4736842105263165e-06,
1052
- "loss": 4.9397,
1053
- "step": 174000
1054
- },
1055
- {
1056
- "epoch": 17.1,
1057
- "learning_rate": 5.263157894736842e-06,
1058
- "loss": 4.9541,
1059
- "step": 175000
1060
- },
1061
- {
1062
- "epoch": 17.2,
1063
- "learning_rate": 5.052631578947369e-06,
1064
- "loss": 4.9449,
1065
- "step": 176000
1066
- },
1067
- {
1068
- "epoch": 17.29,
1069
- "learning_rate": 4.842105263157895e-06,
1070
- "loss": 4.9447,
1071
- "step": 177000
1072
- },
1073
- {
1074
- "epoch": 17.39,
1075
- "learning_rate": 4.631578947368421e-06,
1076
- "loss": 4.9344,
1077
- "step": 178000
1078
- },
1079
- {
1080
- "epoch": 17.49,
1081
- "learning_rate": 4.4210526315789476e-06,
1082
- "loss": 4.933,
1083
- "step": 179000
1084
- },
1085
- {
1086
- "epoch": 17.59,
1087
- "learning_rate": 4.210526315789474e-06,
1088
- "loss": 4.9397,
1089
- "step": 180000
1090
- },
1091
- {
1092
- "epoch": 17.68,
1093
- "learning_rate": 4.000000000000001e-06,
1094
- "loss": 4.9431,
1095
- "step": 181000
1096
- },
1097
- {
1098
- "epoch": 17.78,
1099
- "learning_rate": 3.789473684210527e-06,
1100
- "loss": 4.9378,
1101
- "step": 182000
1102
- },
1103
- {
1104
- "epoch": 17.88,
1105
- "learning_rate": 3.578947368421053e-06,
1106
- "loss": 4.9399,
1107
- "step": 183000
1108
- },
1109
- {
1110
- "epoch": 17.98,
1111
- "learning_rate": 3.368421052631579e-06,
1112
- "loss": 4.9481,
1113
- "step": 184000
1114
- },
1115
- {
1116
- "epoch": 18.08,
1117
- "learning_rate": 3.157894736842105e-06,
1118
- "loss": 4.9399,
1119
- "step": 185000
1120
- },
1121
- {
1122
- "epoch": 18.17,
1123
- "learning_rate": 2.9473684210526317e-06,
1124
- "loss": 4.9365,
1125
- "step": 186000
1126
- },
1127
- {
1128
- "epoch": 18.27,
1129
- "learning_rate": 2.7368421052631583e-06,
1130
- "loss": 4.9307,
1131
- "step": 187000
1132
- },
1133
- {
1134
- "epoch": 18.37,
1135
- "learning_rate": 2.5263157894736844e-06,
1136
- "loss": 4.9391,
1137
- "step": 188000
1138
- },
1139
- {
1140
- "epoch": 18.47,
1141
- "learning_rate": 2.3157894736842105e-06,
1142
- "loss": 4.9331,
1143
- "step": 189000
1144
- },
1145
- {
1146
- "epoch": 18.56,
1147
- "learning_rate": 2.105263157894737e-06,
1148
- "loss": 4.948,
1149
- "step": 190000
1150
- },
1151
- {
1152
- "epoch": 18.66,
1153
- "learning_rate": 1.8947368421052634e-06,
1154
- "loss": 4.9407,
1155
- "step": 191000
1156
- },
1157
- {
1158
- "epoch": 18.76,
1159
- "learning_rate": 1.6842105263157895e-06,
1160
- "loss": 4.944,
1161
- "step": 192000
1162
- },
1163
- {
1164
- "epoch": 18.86,
1165
- "learning_rate": 1.4736842105263159e-06,
1166
- "loss": 4.9286,
1167
- "step": 193000
1168
- }
1169
- ],
1170
- "logging_steps": 1000,
1171
- "max_steps": 200000,
1172
- "num_train_epochs": 20,
1173
- "save_steps": 1000,
1174
- "total_flos": 2.2557198125934313e+19,
1175
- "trial_name": null,
1176
- "trial_params": null
1177
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/checkpoint-193000/training_args.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:c9ff5d6388a59708cb6305a7fab984d438e385b1835680f9629b90c23c5cf704
3
- size 5137
 
 
 
 
len2_5120/checkpoint-194000/config.json DELETED
@@ -1,25 +0,0 @@
1
- {
2
- "architectures": [
3
- "BertForMaskedLM"
4
- ],
5
- "attention_probs_dropout_prob": 0.1,
6
- "classifier_dropout": null,
7
- "hidden_act": "gelu",
8
- "hidden_dropout_prob": 0.1,
9
- "hidden_size": 768,
10
- "initializer_range": 0.02,
11
- "intermediate_size": 3072,
12
- "layer_norm_eps": 1e-12,
13
- "max_length": 512,
14
- "max_position_embeddings": 512,
15
- "model_type": "bert",
16
- "num_attention_heads": 12,
17
- "num_hidden_layers": 12,
18
- "pad_token_id": 0,
19
- "position_embedding_type": "absolute",
20
- "torch_dtype": "float32",
21
- "transformers_version": "4.35.2",
22
- "type_vocab_size": 2,
23
- "use_cache": true,
24
- "vocab_size": 5120
25
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/checkpoint-194000/generation_config.json DELETED
@@ -1,6 +0,0 @@
1
- {
2
- "_from_model_config": true,
3
- "max_length": 512,
4
- "pad_token_id": 0,
5
- "transformers_version": "4.35.2"
6
- }
 
 
 
 
 
 
 
len2_5120/checkpoint-194000/model.safetensors DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:4dddd1badead3e61bee03b327cb7238008ee87237c32a76145e5d7b8cd63c311
3
- size 359944256
 
 
 
 
len2_5120/checkpoint-194000/optimizer.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:bfbb972435c818c69a8a43aae08e74deb9a6222aede1f026af4eefdd90269e25
3
- size 720010699
 
 
 
 
len2_5120/checkpoint-194000/rng_state.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:6960f2857e1f35b7d4b58eb747b95faa2f97f460296f4fb9eeb9499e4792b82e
3
- size 14645
 
 
 
 
len2_5120/checkpoint-194000/scheduler.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:a9aa460464675e9f0593bb3962bdd48c1f52b25c8c8bf257806850efc7689aa1
3
- size 1465
 
 
 
 
len2_5120/checkpoint-194000/trainer_state.json DELETED
@@ -1,1183 +0,0 @@
1
- {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 18.95456765999023,
5
- "eval_steps": 500,
6
- "global_step": 194000,
7
- "is_hyper_param_search": false,
8
- "is_local_process_zero": true,
9
- "is_world_process_zero": true,
10
- "log_history": [
11
- {
12
- "epoch": 0.78,
13
- "learning_rate": 4.000000000000001e-06,
14
- "loss": 7.5832,
15
- "step": 1000
16
- },
17
- {
18
- "epoch": 1.56,
19
- "learning_rate": 8.000000000000001e-06,
20
- "loss": 6.9724,
21
- "step": 2000
22
- },
23
- {
24
- "epoch": 2.34,
25
- "learning_rate": 1.2e-05,
26
- "loss": 6.7695,
27
- "step": 3000
28
- },
29
- {
30
- "epoch": 3.12,
31
- "learning_rate": 1.6000000000000003e-05,
32
- "loss": 6.7002,
33
- "step": 4000
34
- },
35
- {
36
- "epoch": 3.91,
37
- "learning_rate": 2e-05,
38
- "loss": 6.666,
39
- "step": 5000
40
- },
41
- {
42
- "epoch": 4.69,
43
- "learning_rate": 2.4e-05,
44
- "loss": 6.6503,
45
- "step": 6000
46
- },
47
- {
48
- "epoch": 5.47,
49
- "learning_rate": 2.8e-05,
50
- "loss": 6.6407,
51
- "step": 7000
52
- },
53
- {
54
- "epoch": 6.25,
55
- "learning_rate": 3.2000000000000005e-05,
56
- "loss": 6.6277,
57
- "step": 8000
58
- },
59
- {
60
- "epoch": 7.03,
61
- "learning_rate": 3.6e-05,
62
- "loss": 6.6209,
63
- "step": 9000
64
- },
65
- {
66
- "epoch": 7.81,
67
- "learning_rate": 4e-05,
68
- "loss": 6.6135,
69
- "step": 10000
70
- },
71
- {
72
- "epoch": 8.59,
73
- "learning_rate": 3.955555555555556e-05,
74
- "loss": 6.6084,
75
- "step": 11000
76
- },
77
- {
78
- "epoch": 9.38,
79
- "learning_rate": 3.9111111111111115e-05,
80
- "loss": 6.6041,
81
- "step": 12000
82
- },
83
- {
84
- "epoch": 10.16,
85
- "learning_rate": 3.866666666666667e-05,
86
- "loss": 6.6013,
87
- "step": 13000
88
- },
89
- {
90
- "epoch": 10.94,
91
- "learning_rate": 3.8222222222222226e-05,
92
- "loss": 6.5994,
93
- "step": 14000
94
- },
95
- {
96
- "epoch": 11.72,
97
- "learning_rate": 3.777777777777778e-05,
98
- "loss": 6.594,
99
- "step": 15000
100
- },
101
- {
102
- "epoch": 12.5,
103
- "learning_rate": 3.733333333333334e-05,
104
- "loss": 6.5925,
105
- "step": 16000
106
- },
107
- {
108
- "epoch": 13.28,
109
- "learning_rate": 3.6888888888888896e-05,
110
- "loss": 6.5893,
111
- "step": 17000
112
- },
113
- {
114
- "epoch": 14.06,
115
- "learning_rate": 3.644444444444445e-05,
116
- "loss": 6.5875,
117
- "step": 18000
118
- },
119
- {
120
- "epoch": 14.84,
121
- "learning_rate": 3.6e-05,
122
- "loss": 6.5232,
123
- "step": 19000
124
- },
125
- {
126
- "epoch": 15.62,
127
- "learning_rate": 3.555555555555555e-05,
128
- "loss": 6.4358,
129
- "step": 20000
130
- },
131
- {
132
- "epoch": 16.41,
133
- "learning_rate": 3.511111111111111e-05,
134
- "loss": 6.3712,
135
- "step": 21000
136
- },
137
- {
138
- "epoch": 17.19,
139
- "learning_rate": 3.466666666666667e-05,
140
- "loss": 6.311,
141
- "step": 22000
142
- },
143
- {
144
- "epoch": 17.97,
145
- "learning_rate": 3.4222222222222224e-05,
146
- "loss": 6.2493,
147
- "step": 23000
148
- },
149
- {
150
- "epoch": 18.75,
151
- "learning_rate": 3.377777777777778e-05,
152
- "loss": 6.1597,
153
- "step": 24000
154
- },
155
- {
156
- "epoch": 19.53,
157
- "learning_rate": 3.3333333333333335e-05,
158
- "loss": 6.0276,
159
- "step": 25000
160
- },
161
- {
162
- "epoch": 20.31,
163
- "learning_rate": 3.288888888888889e-05,
164
- "loss": 5.8317,
165
- "step": 26000
166
- },
167
- {
168
- "epoch": 21.09,
169
- "learning_rate": 3.2444444444444446e-05,
170
- "loss": 5.6782,
171
- "step": 27000
172
- },
173
- {
174
- "epoch": 21.88,
175
- "learning_rate": 3.2000000000000005e-05,
176
- "loss": 5.5975,
177
- "step": 28000
178
- },
179
- {
180
- "epoch": 22.66,
181
- "learning_rate": 3.155555555555556e-05,
182
- "loss": 5.5369,
183
- "step": 29000
184
- },
185
- {
186
- "epoch": 23.44,
187
- "learning_rate": 3.111111111111112e-05,
188
- "loss": 5.4921,
189
- "step": 30000
190
- },
191
- {
192
- "epoch": 24.22,
193
- "learning_rate": 3.066666666666667e-05,
194
- "loss": 5.4554,
195
- "step": 31000
196
- },
197
- {
198
- "epoch": 25.0,
199
- "learning_rate": 3.0222222222222225e-05,
200
- "loss": 5.4237,
201
- "step": 32000
202
- },
203
- {
204
- "epoch": 25.78,
205
- "learning_rate": 2.977777777777778e-05,
206
- "loss": 5.3938,
207
- "step": 33000
208
- },
209
- {
210
- "epoch": 26.56,
211
- "learning_rate": 2.9333333333333333e-05,
212
- "loss": 5.3701,
213
- "step": 34000
214
- },
215
- {
216
- "epoch": 27.34,
217
- "learning_rate": 2.888888888888889e-05,
218
- "loss": 5.3452,
219
- "step": 35000
220
- },
221
- {
222
- "epoch": 28.12,
223
- "learning_rate": 2.8444444444444447e-05,
224
- "loss": 5.327,
225
- "step": 36000
226
- },
227
- {
228
- "epoch": 28.91,
229
- "learning_rate": 2.8e-05,
230
- "loss": 5.3063,
231
- "step": 37000
232
- },
233
- {
234
- "epoch": 29.69,
235
- "learning_rate": 2.755555555555556e-05,
236
- "loss": 5.2905,
237
- "step": 38000
238
- },
239
- {
240
- "epoch": 30.47,
241
- "learning_rate": 2.7111111111111114e-05,
242
- "loss": 5.2747,
243
- "step": 39000
244
- },
245
- {
246
- "epoch": 31.25,
247
- "learning_rate": 2.6666666666666667e-05,
248
- "loss": 5.2596,
249
- "step": 40000
250
- },
251
- {
252
- "epoch": 32.03,
253
- "learning_rate": 2.6222222222222226e-05,
254
- "loss": 5.2483,
255
- "step": 41000
256
- },
257
- {
258
- "epoch": 32.81,
259
- "learning_rate": 2.577777777777778e-05,
260
- "loss": 5.2336,
261
- "step": 42000
262
- },
263
- {
264
- "epoch": 33.59,
265
- "learning_rate": 2.5333333333333334e-05,
266
- "loss": 5.2214,
267
- "step": 43000
268
- },
269
- {
270
- "epoch": 34.38,
271
- "learning_rate": 2.4888888888888893e-05,
272
- "loss": 5.2103,
273
- "step": 44000
274
- },
275
- {
276
- "epoch": 35.16,
277
- "learning_rate": 2.444444444444445e-05,
278
- "loss": 5.2015,
279
- "step": 45000
280
- },
281
- {
282
- "epoch": 35.94,
283
- "learning_rate": 2.4e-05,
284
- "loss": 5.1907,
285
- "step": 46000
286
- },
287
- {
288
- "epoch": 36.72,
289
- "learning_rate": 2.355555555555556e-05,
290
- "loss": 5.1801,
291
- "step": 47000
292
- },
293
- {
294
- "epoch": 37.5,
295
- "learning_rate": 2.3111111111111112e-05,
296
- "loss": 5.1713,
297
- "step": 48000
298
- },
299
- {
300
- "epoch": 38.28,
301
- "learning_rate": 2.2666666666666668e-05,
302
- "loss": 5.1645,
303
- "step": 49000
304
- },
305
- {
306
- "epoch": 39.06,
307
- "learning_rate": 2.2222222222222227e-05,
308
- "loss": 5.1548,
309
- "step": 50000
310
- },
311
- {
312
- "epoch": 39.84,
313
- "learning_rate": 2.177777777777778e-05,
314
- "loss": 5.1462,
315
- "step": 51000
316
- },
317
- {
318
- "epoch": 40.62,
319
- "learning_rate": 2.1333333333333335e-05,
320
- "loss": 5.1417,
321
- "step": 52000
322
- },
323
- {
324
- "epoch": 41.41,
325
- "learning_rate": 2.088888888888889e-05,
326
- "loss": 5.1305,
327
- "step": 53000
328
- },
329
- {
330
- "epoch": 42.19,
331
- "learning_rate": 2.0444444444444446e-05,
332
- "loss": 5.1287,
333
- "step": 54000
334
- },
335
- {
336
- "epoch": 42.97,
337
- "learning_rate": 2e-05,
338
- "loss": 5.1202,
339
- "step": 55000
340
- },
341
- {
342
- "epoch": 43.75,
343
- "learning_rate": 1.9555555555555557e-05,
344
- "loss": 5.1139,
345
- "step": 56000
346
- },
347
- {
348
- "epoch": 44.53,
349
- "learning_rate": 1.9111111111111113e-05,
350
- "loss": 5.1081,
351
- "step": 57000
352
- },
353
- {
354
- "epoch": 45.31,
355
- "learning_rate": 1.866666666666667e-05,
356
- "loss": 5.1032,
357
- "step": 58000
358
- },
359
- {
360
- "epoch": 46.09,
361
- "learning_rate": 1.8222222222222224e-05,
362
- "loss": 5.0968,
363
- "step": 59000
364
- },
365
- {
366
- "epoch": 46.88,
367
- "learning_rate": 1.7777777777777777e-05,
368
- "loss": 5.0914,
369
- "step": 60000
370
- },
371
- {
372
- "epoch": 47.66,
373
- "learning_rate": 1.7333333333333336e-05,
374
- "loss": 5.085,
375
- "step": 61000
376
- },
377
- {
378
- "epoch": 48.44,
379
- "learning_rate": 1.688888888888889e-05,
380
- "loss": 5.0806,
381
- "step": 62000
382
- },
383
- {
384
- "epoch": 49.22,
385
- "learning_rate": 1.6444444444444444e-05,
386
- "loss": 5.0769,
387
- "step": 63000
388
- },
389
- {
390
- "epoch": 50.0,
391
- "learning_rate": 1.6000000000000003e-05,
392
- "loss": 5.0728,
393
- "step": 64000
394
- },
395
- {
396
- "epoch": 50.78,
397
- "learning_rate": 1.555555555555556e-05,
398
- "loss": 5.0692,
399
- "step": 65000
400
- },
401
- {
402
- "epoch": 51.56,
403
- "learning_rate": 1.5111111111111112e-05,
404
- "loss": 5.065,
405
- "step": 66000
406
- },
407
- {
408
- "epoch": 52.34,
409
- "learning_rate": 1.4666666666666666e-05,
410
- "loss": 5.0594,
411
- "step": 67000
412
- },
413
- {
414
- "epoch": 53.12,
415
- "learning_rate": 1.4222222222222224e-05,
416
- "loss": 5.0549,
417
- "step": 68000
418
- },
419
- {
420
- "epoch": 53.91,
421
- "learning_rate": 1.377777777777778e-05,
422
- "loss": 5.0538,
423
- "step": 69000
424
- },
425
- {
426
- "epoch": 54.69,
427
- "learning_rate": 1.3333333333333333e-05,
428
- "loss": 5.0482,
429
- "step": 70000
430
- },
431
- {
432
- "epoch": 55.47,
433
- "learning_rate": 1.288888888888889e-05,
434
- "loss": 5.0475,
435
- "step": 71000
436
- },
437
- {
438
- "epoch": 56.25,
439
- "learning_rate": 1.2444444444444446e-05,
440
- "loss": 5.0399,
441
- "step": 72000
442
- },
443
- {
444
- "epoch": 57.03,
445
- "learning_rate": 1.2e-05,
446
- "loss": 5.0406,
447
- "step": 73000
448
- },
449
- {
450
- "epoch": 57.81,
451
- "learning_rate": 1.1555555555555556e-05,
452
- "loss": 5.037,
453
- "step": 74000
454
- },
455
- {
456
- "epoch": 58.59,
457
- "learning_rate": 1.1111111111111113e-05,
458
- "loss": 5.0326,
459
- "step": 75000
460
- },
461
- {
462
- "epoch": 59.38,
463
- "learning_rate": 1.0666666666666667e-05,
464
- "loss": 5.031,
465
- "step": 76000
466
- },
467
- {
468
- "epoch": 60.16,
469
- "learning_rate": 1.0222222222222223e-05,
470
- "loss": 5.0291,
471
- "step": 77000
472
- },
473
- {
474
- "epoch": 60.94,
475
- "learning_rate": 9.777777777777779e-06,
476
- "loss": 5.0266,
477
- "step": 78000
478
- },
479
- {
480
- "epoch": 61.72,
481
- "learning_rate": 9.333333333333334e-06,
482
- "loss": 5.0235,
483
- "step": 79000
484
- },
485
- {
486
- "epoch": 62.5,
487
- "learning_rate": 8.888888888888888e-06,
488
- "loss": 5.0211,
489
- "step": 80000
490
- },
491
- {
492
- "epoch": 63.28,
493
- "learning_rate": 8.444444444444446e-06,
494
- "loss": 5.0214,
495
- "step": 81000
496
- },
497
- {
498
- "epoch": 64.06,
499
- "learning_rate": 8.000000000000001e-06,
500
- "loss": 5.0176,
501
- "step": 82000
502
- },
503
- {
504
- "epoch": 64.84,
505
- "learning_rate": 7.555555555555556e-06,
506
- "loss": 5.0157,
507
- "step": 83000
508
- },
509
- {
510
- "epoch": 65.62,
511
- "learning_rate": 7.111111111111112e-06,
512
- "loss": 5.014,
513
- "step": 84000
514
- },
515
- {
516
- "epoch": 66.41,
517
- "learning_rate": 6.666666666666667e-06,
518
- "loss": 5.0115,
519
- "step": 85000
520
- },
521
- {
522
- "epoch": 67.19,
523
- "learning_rate": 6.222222222222223e-06,
524
- "loss": 5.0087,
525
- "step": 86000
526
- },
527
- {
528
- "epoch": 67.97,
529
- "learning_rate": 5.777777777777778e-06,
530
- "loss": 5.0105,
531
- "step": 87000
532
- },
533
- {
534
- "epoch": 68.75,
535
- "learning_rate": 5.333333333333334e-06,
536
- "loss": 5.0062,
537
- "step": 88000
538
- },
539
- {
540
- "epoch": 69.53,
541
- "learning_rate": 4.888888888888889e-06,
542
- "loss": 5.0028,
543
- "step": 89000
544
- },
545
- {
546
- "epoch": 70.31,
547
- "learning_rate": 4.444444444444444e-06,
548
- "loss": 5.0066,
549
- "step": 90000
550
- },
551
- {
552
- "epoch": 71.09,
553
- "learning_rate": 4.000000000000001e-06,
554
- "loss": 5.0032,
555
- "step": 91000
556
- },
557
- {
558
- "epoch": 71.88,
559
- "learning_rate": 3.555555555555556e-06,
560
- "loss": 5.0022,
561
- "step": 92000
562
- },
563
- {
564
- "epoch": 72.66,
565
- "learning_rate": 3.1111111111111116e-06,
566
- "loss": 5.0024,
567
- "step": 93000
568
- },
569
- {
570
- "epoch": 73.44,
571
- "learning_rate": 2.666666666666667e-06,
572
- "loss": 5.0007,
573
- "step": 94000
574
- },
575
- {
576
- "epoch": 74.22,
577
- "learning_rate": 2.222222222222222e-06,
578
- "loss": 5.0008,
579
- "step": 95000
580
- },
581
- {
582
- "epoch": 75.0,
583
- "learning_rate": 1.777777777777778e-06,
584
- "loss": 4.9978,
585
- "step": 96000
586
- },
587
- {
588
- "epoch": 75.78,
589
- "learning_rate": 1.3333333333333334e-06,
590
- "loss": 4.9966,
591
- "step": 97000
592
- },
593
- {
594
- "epoch": 76.56,
595
- "learning_rate": 8.88888888888889e-07,
596
- "loss": 4.9994,
597
- "step": 98000
598
- },
599
- {
600
- "epoch": 77.34,
601
- "learning_rate": 4.444444444444445e-07,
602
- "loss": 4.9966,
603
- "step": 99000
604
- },
605
- {
606
- "epoch": 78.12,
607
- "learning_rate": 0.0,
608
- "loss": 4.9976,
609
- "step": 100000
610
- },
611
- {
612
- "epoch": 9.87,
613
- "learning_rate": 2.0842105263157895e-05,
614
- "loss": 5.0155,
615
- "step": 101000
616
- },
617
- {
618
- "epoch": 9.97,
619
- "learning_rate": 2.0631578947368425e-05,
620
- "loss": 5.024,
621
- "step": 102000
622
- },
623
- {
624
- "epoch": 10.06,
625
- "learning_rate": 2.0421052631578952e-05,
626
- "loss": 5.0241,
627
- "step": 103000
628
- },
629
- {
630
- "epoch": 10.16,
631
- "learning_rate": 2.0210526315789475e-05,
632
- "loss": 5.0218,
633
- "step": 104000
634
- },
635
- {
636
- "epoch": 10.26,
637
- "learning_rate": 2e-05,
638
- "loss": 5.0231,
639
- "step": 105000
640
- },
641
- {
642
- "epoch": 10.36,
643
- "learning_rate": 1.9789473684210528e-05,
644
- "loss": 5.0231,
645
- "step": 106000
646
- },
647
- {
648
- "epoch": 10.45,
649
- "learning_rate": 1.9578947368421055e-05,
650
- "loss": 5.0198,
651
- "step": 107000
652
- },
653
- {
654
- "epoch": 10.55,
655
- "learning_rate": 1.936842105263158e-05,
656
- "loss": 5.0158,
657
- "step": 108000
658
- },
659
- {
660
- "epoch": 10.65,
661
- "learning_rate": 1.9157894736842108e-05,
662
- "loss": 5.0122,
663
- "step": 109000
664
- },
665
- {
666
- "epoch": 10.75,
667
- "learning_rate": 1.894736842105263e-05,
668
- "loss": 5.0165,
669
- "step": 110000
670
- },
671
- {
672
- "epoch": 10.85,
673
- "learning_rate": 1.873684210526316e-05,
674
- "loss": 5.0183,
675
- "step": 111000
676
- },
677
- {
678
- "epoch": 10.94,
679
- "learning_rate": 1.8526315789473684e-05,
680
- "loss": 5.01,
681
- "step": 112000
682
- },
683
- {
684
- "epoch": 11.04,
685
- "learning_rate": 1.831578947368421e-05,
686
- "loss": 5.007,
687
- "step": 113000
688
- },
689
- {
690
- "epoch": 11.14,
691
- "learning_rate": 1.810526315789474e-05,
692
- "loss": 5.0059,
693
- "step": 114000
694
- },
695
- {
696
- "epoch": 11.24,
697
- "learning_rate": 1.7894736842105264e-05,
698
- "loss": 5.0089,
699
- "step": 115000
700
- },
701
- {
702
- "epoch": 11.33,
703
- "learning_rate": 1.768421052631579e-05,
704
- "loss": 5.0081,
705
- "step": 116000
706
- },
707
- {
708
- "epoch": 11.43,
709
- "learning_rate": 1.7473684210526317e-05,
710
- "loss": 5.002,
711
- "step": 117000
712
- },
713
- {
714
- "epoch": 11.53,
715
- "learning_rate": 1.7263157894736843e-05,
716
- "loss": 5.0052,
717
- "step": 118000
718
- },
719
- {
720
- "epoch": 11.63,
721
- "learning_rate": 1.705263157894737e-05,
722
- "loss": 5.0079,
723
- "step": 119000
724
- },
725
- {
726
- "epoch": 11.72,
727
- "learning_rate": 1.6842105263157896e-05,
728
- "loss": 4.9985,
729
- "step": 120000
730
- },
731
- {
732
- "epoch": 11.82,
733
- "learning_rate": 1.6631578947368423e-05,
734
- "loss": 5.006,
735
- "step": 121000
736
- },
737
- {
738
- "epoch": 11.92,
739
- "learning_rate": 1.642105263157895e-05,
740
- "loss": 4.9956,
741
- "step": 122000
742
- },
743
- {
744
- "epoch": 12.02,
745
- "learning_rate": 1.6210526315789473e-05,
746
- "loss": 4.9973,
747
- "step": 123000
748
- },
749
- {
750
- "epoch": 12.12,
751
- "learning_rate": 1.6000000000000003e-05,
752
- "loss": 4.9976,
753
- "step": 124000
754
- },
755
- {
756
- "epoch": 12.21,
757
- "learning_rate": 1.578947368421053e-05,
758
- "loss": 4.9882,
759
- "step": 125000
760
- },
761
- {
762
- "epoch": 12.31,
763
- "learning_rate": 1.5578947368421052e-05,
764
- "loss": 4.9908,
765
- "step": 126000
766
- },
767
- {
768
- "epoch": 12.41,
769
- "learning_rate": 1.536842105263158e-05,
770
- "loss": 5.0038,
771
- "step": 127000
772
- },
773
- {
774
- "epoch": 12.51,
775
- "learning_rate": 1.5157894736842107e-05,
776
- "loss": 4.9923,
777
- "step": 128000
778
- },
779
- {
780
- "epoch": 12.6,
781
- "learning_rate": 1.4947368421052632e-05,
782
- "loss": 4.9847,
783
- "step": 129000
784
- },
785
- {
786
- "epoch": 12.7,
787
- "learning_rate": 1.4736842105263159e-05,
788
- "loss": 4.9864,
789
- "step": 130000
790
- },
791
- {
792
- "epoch": 12.8,
793
- "learning_rate": 1.4526315789473687e-05,
794
- "loss": 4.9903,
795
- "step": 131000
796
- },
797
- {
798
- "epoch": 12.9,
799
- "learning_rate": 1.4315789473684212e-05,
800
- "loss": 4.9812,
801
- "step": 132000
802
- },
803
- {
804
- "epoch": 12.99,
805
- "learning_rate": 1.4105263157894738e-05,
806
- "loss": 4.9923,
807
- "step": 133000
808
- },
809
- {
810
- "epoch": 13.09,
811
- "learning_rate": 1.3894736842105265e-05,
812
- "loss": 4.9838,
813
- "step": 134000
814
- },
815
- {
816
- "epoch": 13.19,
817
- "learning_rate": 1.3684210526315791e-05,
818
- "loss": 4.977,
819
- "step": 135000
820
- },
821
- {
822
- "epoch": 13.29,
823
- "learning_rate": 1.3473684210526316e-05,
824
- "loss": 4.9743,
825
- "step": 136000
826
- },
827
- {
828
- "epoch": 13.39,
829
- "learning_rate": 1.3263157894736843e-05,
830
- "loss": 4.9797,
831
- "step": 137000
832
- },
833
- {
834
- "epoch": 13.48,
835
- "learning_rate": 1.305263157894737e-05,
836
- "loss": 4.9915,
837
- "step": 138000
838
- },
839
- {
840
- "epoch": 13.58,
841
- "learning_rate": 1.2842105263157896e-05,
842
- "loss": 4.9762,
843
- "step": 139000
844
- },
845
- {
846
- "epoch": 13.68,
847
- "learning_rate": 1.263157894736842e-05,
848
- "loss": 4.9771,
849
- "step": 140000
850
- },
851
- {
852
- "epoch": 13.78,
853
- "learning_rate": 1.2421052631578949e-05,
854
- "loss": 4.9763,
855
- "step": 141000
856
- },
857
- {
858
- "epoch": 13.87,
859
- "learning_rate": 1.2210526315789475e-05,
860
- "loss": 4.9706,
861
- "step": 142000
862
- },
863
- {
864
- "epoch": 13.97,
865
- "learning_rate": 1.2e-05,
866
- "loss": 4.9731,
867
- "step": 143000
868
- },
869
- {
870
- "epoch": 14.07,
871
- "learning_rate": 1.1789473684210527e-05,
872
- "loss": 4.9736,
873
- "step": 144000
874
- },
875
- {
876
- "epoch": 14.17,
877
- "learning_rate": 1.1578947368421053e-05,
878
- "loss": 4.9704,
879
- "step": 145000
880
- },
881
- {
882
- "epoch": 14.26,
883
- "learning_rate": 1.136842105263158e-05,
884
- "loss": 4.9762,
885
- "step": 146000
886
- },
887
- {
888
- "epoch": 14.36,
889
- "learning_rate": 1.1157894736842105e-05,
890
- "loss": 4.9709,
891
- "step": 147000
892
- },
893
- {
894
- "epoch": 14.46,
895
- "learning_rate": 1.0947368421052633e-05,
896
- "loss": 4.957,
897
- "step": 148000
898
- },
899
- {
900
- "epoch": 14.56,
901
- "learning_rate": 1.073684210526316e-05,
902
- "loss": 4.9666,
903
- "step": 149000
904
- },
905
- {
906
- "epoch": 14.66,
907
- "learning_rate": 1.0526315789473684e-05,
908
- "loss": 4.9623,
909
- "step": 150000
910
- },
911
- {
912
- "epoch": 14.75,
913
- "learning_rate": 1.0315789473684213e-05,
914
- "loss": 4.9613,
915
- "step": 151000
916
- },
917
- {
918
- "epoch": 14.85,
919
- "learning_rate": 1.0105263157894738e-05,
920
- "loss": 4.9642,
921
- "step": 152000
922
- },
923
- {
924
- "epoch": 14.95,
925
- "learning_rate": 9.894736842105264e-06,
926
- "loss": 4.9605,
927
- "step": 153000
928
- },
929
- {
930
- "epoch": 15.05,
931
- "learning_rate": 9.68421052631579e-06,
932
- "loss": 4.9688,
933
- "step": 154000
934
- },
935
- {
936
- "epoch": 15.14,
937
- "learning_rate": 9.473684210526315e-06,
938
- "loss": 4.9635,
939
- "step": 155000
940
- },
941
- {
942
- "epoch": 15.24,
943
- "learning_rate": 9.263157894736842e-06,
944
- "loss": 4.9587,
945
- "step": 156000
946
- },
947
- {
948
- "epoch": 15.34,
949
- "learning_rate": 9.05263157894737e-06,
950
- "loss": 4.9633,
951
- "step": 157000
952
- },
953
- {
954
- "epoch": 15.44,
955
- "learning_rate": 8.842105263157895e-06,
956
- "loss": 4.9613,
957
- "step": 158000
958
- },
959
- {
960
- "epoch": 15.53,
961
- "learning_rate": 8.631578947368422e-06,
962
- "loss": 4.9629,
963
- "step": 159000
964
- },
965
- {
966
- "epoch": 15.63,
967
- "learning_rate": 8.421052631578948e-06,
968
- "loss": 4.9547,
969
- "step": 160000
970
- },
971
- {
972
- "epoch": 15.73,
973
- "learning_rate": 8.210526315789475e-06,
974
- "loss": 4.9541,
975
- "step": 161000
976
- },
977
- {
978
- "epoch": 15.83,
979
- "learning_rate": 8.000000000000001e-06,
980
- "loss": 4.9532,
981
- "step": 162000
982
- },
983
- {
984
- "epoch": 15.93,
985
- "learning_rate": 7.789473684210526e-06,
986
- "loss": 4.9461,
987
- "step": 163000
988
- },
989
- {
990
- "epoch": 16.02,
991
- "learning_rate": 7.578947368421054e-06,
992
- "loss": 4.9518,
993
- "step": 164000
994
- },
995
- {
996
- "epoch": 16.12,
997
- "learning_rate": 7.368421052631579e-06,
998
- "loss": 4.9543,
999
- "step": 165000
1000
- },
1001
- {
1002
- "epoch": 16.22,
1003
- "learning_rate": 7.157894736842106e-06,
1004
- "loss": 4.9526,
1005
- "step": 166000
1006
- },
1007
- {
1008
- "epoch": 16.32,
1009
- "learning_rate": 6.947368421052632e-06,
1010
- "loss": 4.9447,
1011
- "step": 167000
1012
- },
1013
- {
1014
- "epoch": 16.41,
1015
- "learning_rate": 6.736842105263158e-06,
1016
- "loss": 4.9545,
1017
- "step": 168000
1018
- },
1019
- {
1020
- "epoch": 16.51,
1021
- "learning_rate": 6.526315789473685e-06,
1022
- "loss": 4.9448,
1023
- "step": 169000
1024
- },
1025
- {
1026
- "epoch": 16.61,
1027
- "learning_rate": 6.31578947368421e-06,
1028
- "loss": 4.9521,
1029
- "step": 170000
1030
- },
1031
- {
1032
- "epoch": 16.71,
1033
- "learning_rate": 6.105263157894738e-06,
1034
- "loss": 4.9503,
1035
- "step": 171000
1036
- },
1037
- {
1038
- "epoch": 16.81,
1039
- "learning_rate": 5.8947368421052634e-06,
1040
- "loss": 4.9537,
1041
- "step": 172000
1042
- },
1043
- {
1044
- "epoch": 16.9,
1045
- "learning_rate": 5.68421052631579e-06,
1046
- "loss": 4.9446,
1047
- "step": 173000
1048
- },
1049
- {
1050
- "epoch": 17.0,
1051
- "learning_rate": 5.4736842105263165e-06,
1052
- "loss": 4.9397,
1053
- "step": 174000
1054
- },
1055
- {
1056
- "epoch": 17.1,
1057
- "learning_rate": 5.263157894736842e-06,
1058
- "loss": 4.9541,
1059
- "step": 175000
1060
- },
1061
- {
1062
- "epoch": 17.2,
1063
- "learning_rate": 5.052631578947369e-06,
1064
- "loss": 4.9449,
1065
- "step": 176000
1066
- },
1067
- {
1068
- "epoch": 17.29,
1069
- "learning_rate": 4.842105263157895e-06,
1070
- "loss": 4.9447,
1071
- "step": 177000
1072
- },
1073
- {
1074
- "epoch": 17.39,
1075
- "learning_rate": 4.631578947368421e-06,
1076
- "loss": 4.9344,
1077
- "step": 178000
1078
- },
1079
- {
1080
- "epoch": 17.49,
1081
- "learning_rate": 4.4210526315789476e-06,
1082
- "loss": 4.933,
1083
- "step": 179000
1084
- },
1085
- {
1086
- "epoch": 17.59,
1087
- "learning_rate": 4.210526315789474e-06,
1088
- "loss": 4.9397,
1089
- "step": 180000
1090
- },
1091
- {
1092
- "epoch": 17.68,
1093
- "learning_rate": 4.000000000000001e-06,
1094
- "loss": 4.9431,
1095
- "step": 181000
1096
- },
1097
- {
1098
- "epoch": 17.78,
1099
- "learning_rate": 3.789473684210527e-06,
1100
- "loss": 4.9378,
1101
- "step": 182000
1102
- },
1103
- {
1104
- "epoch": 17.88,
1105
- "learning_rate": 3.578947368421053e-06,
1106
- "loss": 4.9399,
1107
- "step": 183000
1108
- },
1109
- {
1110
- "epoch": 17.98,
1111
- "learning_rate": 3.368421052631579e-06,
1112
- "loss": 4.9481,
1113
- "step": 184000
1114
- },
1115
- {
1116
- "epoch": 18.08,
1117
- "learning_rate": 3.157894736842105e-06,
1118
- "loss": 4.9399,
1119
- "step": 185000
1120
- },
1121
- {
1122
- "epoch": 18.17,
1123
- "learning_rate": 2.9473684210526317e-06,
1124
- "loss": 4.9365,
1125
- "step": 186000
1126
- },
1127
- {
1128
- "epoch": 18.27,
1129
- "learning_rate": 2.7368421052631583e-06,
1130
- "loss": 4.9307,
1131
- "step": 187000
1132
- },
1133
- {
1134
- "epoch": 18.37,
1135
- "learning_rate": 2.5263157894736844e-06,
1136
- "loss": 4.9391,
1137
- "step": 188000
1138
- },
1139
- {
1140
- "epoch": 18.47,
1141
- "learning_rate": 2.3157894736842105e-06,
1142
- "loss": 4.9331,
1143
- "step": 189000
1144
- },
1145
- {
1146
- "epoch": 18.56,
1147
- "learning_rate": 2.105263157894737e-06,
1148
- "loss": 4.948,
1149
- "step": 190000
1150
- },
1151
- {
1152
- "epoch": 18.66,
1153
- "learning_rate": 1.8947368421052634e-06,
1154
- "loss": 4.9407,
1155
- "step": 191000
1156
- },
1157
- {
1158
- "epoch": 18.76,
1159
- "learning_rate": 1.6842105263157895e-06,
1160
- "loss": 4.944,
1161
- "step": 192000
1162
- },
1163
- {
1164
- "epoch": 18.86,
1165
- "learning_rate": 1.4736842105263159e-06,
1166
- "loss": 4.9286,
1167
- "step": 193000
1168
- },
1169
- {
1170
- "epoch": 18.95,
1171
- "learning_rate": 1.2631578947368422e-06,
1172
- "loss": 4.9338,
1173
- "step": 194000
1174
- }
1175
- ],
1176
- "logging_steps": 1000,
1177
- "max_steps": 200000,
1178
- "num_train_epochs": 20,
1179
- "save_steps": 1000,
1180
- "total_flos": 2.258245829660849e+19,
1181
- "trial_name": null,
1182
- "trial_params": null
1183
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/checkpoint-194000/training_args.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:c9ff5d6388a59708cb6305a7fab984d438e385b1835680f9629b90c23c5cf704
3
- size 5137
 
 
 
 
len2_5120/checkpoint-195000/config.json DELETED
@@ -1,25 +0,0 @@
1
- {
2
- "architectures": [
3
- "BertForMaskedLM"
4
- ],
5
- "attention_probs_dropout_prob": 0.1,
6
- "classifier_dropout": null,
7
- "hidden_act": "gelu",
8
- "hidden_dropout_prob": 0.1,
9
- "hidden_size": 768,
10
- "initializer_range": 0.02,
11
- "intermediate_size": 3072,
12
- "layer_norm_eps": 1e-12,
13
- "max_length": 512,
14
- "max_position_embeddings": 512,
15
- "model_type": "bert",
16
- "num_attention_heads": 12,
17
- "num_hidden_layers": 12,
18
- "pad_token_id": 0,
19
- "position_embedding_type": "absolute",
20
- "torch_dtype": "float32",
21
- "transformers_version": "4.35.2",
22
- "type_vocab_size": 2,
23
- "use_cache": true,
24
- "vocab_size": 5120
25
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/checkpoint-195000/generation_config.json DELETED
@@ -1,6 +0,0 @@
1
- {
2
- "_from_model_config": true,
3
- "max_length": 512,
4
- "pad_token_id": 0,
5
- "transformers_version": "4.35.2"
6
- }
 
 
 
 
 
 
 
len2_5120/checkpoint-195000/model.safetensors DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:f08acae5c42aec4d08dd4c07ecf0003d0906dac4b6898a7a28eaa17d433b6b37
3
- size 359944256
 
 
 
 
len2_5120/checkpoint-195000/optimizer.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:557e5f514efd168ec0a2e047f5f68ef4d0d9419d2589e23428e517eb9b2fcc41
3
- size 720010699
 
 
 
 
len2_5120/checkpoint-195000/rng_state.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:4c417c0f604cac9e266875d12cb57ca16657d4c563a05f5a5e3e6ef9004ab158
3
- size 14645
 
 
 
 
len2_5120/checkpoint-195000/scheduler.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:ac3001d5dd025a8da6bc02910e07eb74e4b31d248926c79f2143788adccb7d21
3
- size 1465
 
 
 
 
len2_5120/checkpoint-195000/trainer_state.json DELETED
@@ -1,1189 +0,0 @@
1
- {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 19.05227161700049,
5
- "eval_steps": 500,
6
- "global_step": 195000,
7
- "is_hyper_param_search": false,
8
- "is_local_process_zero": true,
9
- "is_world_process_zero": true,
10
- "log_history": [
11
- {
12
- "epoch": 0.78,
13
- "learning_rate": 4.000000000000001e-06,
14
- "loss": 7.5832,
15
- "step": 1000
16
- },
17
- {
18
- "epoch": 1.56,
19
- "learning_rate": 8.000000000000001e-06,
20
- "loss": 6.9724,
21
- "step": 2000
22
- },
23
- {
24
- "epoch": 2.34,
25
- "learning_rate": 1.2e-05,
26
- "loss": 6.7695,
27
- "step": 3000
28
- },
29
- {
30
- "epoch": 3.12,
31
- "learning_rate": 1.6000000000000003e-05,
32
- "loss": 6.7002,
33
- "step": 4000
34
- },
35
- {
36
- "epoch": 3.91,
37
- "learning_rate": 2e-05,
38
- "loss": 6.666,
39
- "step": 5000
40
- },
41
- {
42
- "epoch": 4.69,
43
- "learning_rate": 2.4e-05,
44
- "loss": 6.6503,
45
- "step": 6000
46
- },
47
- {
48
- "epoch": 5.47,
49
- "learning_rate": 2.8e-05,
50
- "loss": 6.6407,
51
- "step": 7000
52
- },
53
- {
54
- "epoch": 6.25,
55
- "learning_rate": 3.2000000000000005e-05,
56
- "loss": 6.6277,
57
- "step": 8000
58
- },
59
- {
60
- "epoch": 7.03,
61
- "learning_rate": 3.6e-05,
62
- "loss": 6.6209,
63
- "step": 9000
64
- },
65
- {
66
- "epoch": 7.81,
67
- "learning_rate": 4e-05,
68
- "loss": 6.6135,
69
- "step": 10000
70
- },
71
- {
72
- "epoch": 8.59,
73
- "learning_rate": 3.955555555555556e-05,
74
- "loss": 6.6084,
75
- "step": 11000
76
- },
77
- {
78
- "epoch": 9.38,
79
- "learning_rate": 3.9111111111111115e-05,
80
- "loss": 6.6041,
81
- "step": 12000
82
- },
83
- {
84
- "epoch": 10.16,
85
- "learning_rate": 3.866666666666667e-05,
86
- "loss": 6.6013,
87
- "step": 13000
88
- },
89
- {
90
- "epoch": 10.94,
91
- "learning_rate": 3.8222222222222226e-05,
92
- "loss": 6.5994,
93
- "step": 14000
94
- },
95
- {
96
- "epoch": 11.72,
97
- "learning_rate": 3.777777777777778e-05,
98
- "loss": 6.594,
99
- "step": 15000
100
- },
101
- {
102
- "epoch": 12.5,
103
- "learning_rate": 3.733333333333334e-05,
104
- "loss": 6.5925,
105
- "step": 16000
106
- },
107
- {
108
- "epoch": 13.28,
109
- "learning_rate": 3.6888888888888896e-05,
110
- "loss": 6.5893,
111
- "step": 17000
112
- },
113
- {
114
- "epoch": 14.06,
115
- "learning_rate": 3.644444444444445e-05,
116
- "loss": 6.5875,
117
- "step": 18000
118
- },
119
- {
120
- "epoch": 14.84,
121
- "learning_rate": 3.6e-05,
122
- "loss": 6.5232,
123
- "step": 19000
124
- },
125
- {
126
- "epoch": 15.62,
127
- "learning_rate": 3.555555555555555e-05,
128
- "loss": 6.4358,
129
- "step": 20000
130
- },
131
- {
132
- "epoch": 16.41,
133
- "learning_rate": 3.511111111111111e-05,
134
- "loss": 6.3712,
135
- "step": 21000
136
- },
137
- {
138
- "epoch": 17.19,
139
- "learning_rate": 3.466666666666667e-05,
140
- "loss": 6.311,
141
- "step": 22000
142
- },
143
- {
144
- "epoch": 17.97,
145
- "learning_rate": 3.4222222222222224e-05,
146
- "loss": 6.2493,
147
- "step": 23000
148
- },
149
- {
150
- "epoch": 18.75,
151
- "learning_rate": 3.377777777777778e-05,
152
- "loss": 6.1597,
153
- "step": 24000
154
- },
155
- {
156
- "epoch": 19.53,
157
- "learning_rate": 3.3333333333333335e-05,
158
- "loss": 6.0276,
159
- "step": 25000
160
- },
161
- {
162
- "epoch": 20.31,
163
- "learning_rate": 3.288888888888889e-05,
164
- "loss": 5.8317,
165
- "step": 26000
166
- },
167
- {
168
- "epoch": 21.09,
169
- "learning_rate": 3.2444444444444446e-05,
170
- "loss": 5.6782,
171
- "step": 27000
172
- },
173
- {
174
- "epoch": 21.88,
175
- "learning_rate": 3.2000000000000005e-05,
176
- "loss": 5.5975,
177
- "step": 28000
178
- },
179
- {
180
- "epoch": 22.66,
181
- "learning_rate": 3.155555555555556e-05,
182
- "loss": 5.5369,
183
- "step": 29000
184
- },
185
- {
186
- "epoch": 23.44,
187
- "learning_rate": 3.111111111111112e-05,
188
- "loss": 5.4921,
189
- "step": 30000
190
- },
191
- {
192
- "epoch": 24.22,
193
- "learning_rate": 3.066666666666667e-05,
194
- "loss": 5.4554,
195
- "step": 31000
196
- },
197
- {
198
- "epoch": 25.0,
199
- "learning_rate": 3.0222222222222225e-05,
200
- "loss": 5.4237,
201
- "step": 32000
202
- },
203
- {
204
- "epoch": 25.78,
205
- "learning_rate": 2.977777777777778e-05,
206
- "loss": 5.3938,
207
- "step": 33000
208
- },
209
- {
210
- "epoch": 26.56,
211
- "learning_rate": 2.9333333333333333e-05,
212
- "loss": 5.3701,
213
- "step": 34000
214
- },
215
- {
216
- "epoch": 27.34,
217
- "learning_rate": 2.888888888888889e-05,
218
- "loss": 5.3452,
219
- "step": 35000
220
- },
221
- {
222
- "epoch": 28.12,
223
- "learning_rate": 2.8444444444444447e-05,
224
- "loss": 5.327,
225
- "step": 36000
226
- },
227
- {
228
- "epoch": 28.91,
229
- "learning_rate": 2.8e-05,
230
- "loss": 5.3063,
231
- "step": 37000
232
- },
233
- {
234
- "epoch": 29.69,
235
- "learning_rate": 2.755555555555556e-05,
236
- "loss": 5.2905,
237
- "step": 38000
238
- },
239
- {
240
- "epoch": 30.47,
241
- "learning_rate": 2.7111111111111114e-05,
242
- "loss": 5.2747,
243
- "step": 39000
244
- },
245
- {
246
- "epoch": 31.25,
247
- "learning_rate": 2.6666666666666667e-05,
248
- "loss": 5.2596,
249
- "step": 40000
250
- },
251
- {
252
- "epoch": 32.03,
253
- "learning_rate": 2.6222222222222226e-05,
254
- "loss": 5.2483,
255
- "step": 41000
256
- },
257
- {
258
- "epoch": 32.81,
259
- "learning_rate": 2.577777777777778e-05,
260
- "loss": 5.2336,
261
- "step": 42000
262
- },
263
- {
264
- "epoch": 33.59,
265
- "learning_rate": 2.5333333333333334e-05,
266
- "loss": 5.2214,
267
- "step": 43000
268
- },
269
- {
270
- "epoch": 34.38,
271
- "learning_rate": 2.4888888888888893e-05,
272
- "loss": 5.2103,
273
- "step": 44000
274
- },
275
- {
276
- "epoch": 35.16,
277
- "learning_rate": 2.444444444444445e-05,
278
- "loss": 5.2015,
279
- "step": 45000
280
- },
281
- {
282
- "epoch": 35.94,
283
- "learning_rate": 2.4e-05,
284
- "loss": 5.1907,
285
- "step": 46000
286
- },
287
- {
288
- "epoch": 36.72,
289
- "learning_rate": 2.355555555555556e-05,
290
- "loss": 5.1801,
291
- "step": 47000
292
- },
293
- {
294
- "epoch": 37.5,
295
- "learning_rate": 2.3111111111111112e-05,
296
- "loss": 5.1713,
297
- "step": 48000
298
- },
299
- {
300
- "epoch": 38.28,
301
- "learning_rate": 2.2666666666666668e-05,
302
- "loss": 5.1645,
303
- "step": 49000
304
- },
305
- {
306
- "epoch": 39.06,
307
- "learning_rate": 2.2222222222222227e-05,
308
- "loss": 5.1548,
309
- "step": 50000
310
- },
311
- {
312
- "epoch": 39.84,
313
- "learning_rate": 2.177777777777778e-05,
314
- "loss": 5.1462,
315
- "step": 51000
316
- },
317
- {
318
- "epoch": 40.62,
319
- "learning_rate": 2.1333333333333335e-05,
320
- "loss": 5.1417,
321
- "step": 52000
322
- },
323
- {
324
- "epoch": 41.41,
325
- "learning_rate": 2.088888888888889e-05,
326
- "loss": 5.1305,
327
- "step": 53000
328
- },
329
- {
330
- "epoch": 42.19,
331
- "learning_rate": 2.0444444444444446e-05,
332
- "loss": 5.1287,
333
- "step": 54000
334
- },
335
- {
336
- "epoch": 42.97,
337
- "learning_rate": 2e-05,
338
- "loss": 5.1202,
339
- "step": 55000
340
- },
341
- {
342
- "epoch": 43.75,
343
- "learning_rate": 1.9555555555555557e-05,
344
- "loss": 5.1139,
345
- "step": 56000
346
- },
347
- {
348
- "epoch": 44.53,
349
- "learning_rate": 1.9111111111111113e-05,
350
- "loss": 5.1081,
351
- "step": 57000
352
- },
353
- {
354
- "epoch": 45.31,
355
- "learning_rate": 1.866666666666667e-05,
356
- "loss": 5.1032,
357
- "step": 58000
358
- },
359
- {
360
- "epoch": 46.09,
361
- "learning_rate": 1.8222222222222224e-05,
362
- "loss": 5.0968,
363
- "step": 59000
364
- },
365
- {
366
- "epoch": 46.88,
367
- "learning_rate": 1.7777777777777777e-05,
368
- "loss": 5.0914,
369
- "step": 60000
370
- },
371
- {
372
- "epoch": 47.66,
373
- "learning_rate": 1.7333333333333336e-05,
374
- "loss": 5.085,
375
- "step": 61000
376
- },
377
- {
378
- "epoch": 48.44,
379
- "learning_rate": 1.688888888888889e-05,
380
- "loss": 5.0806,
381
- "step": 62000
382
- },
383
- {
384
- "epoch": 49.22,
385
- "learning_rate": 1.6444444444444444e-05,
386
- "loss": 5.0769,
387
- "step": 63000
388
- },
389
- {
390
- "epoch": 50.0,
391
- "learning_rate": 1.6000000000000003e-05,
392
- "loss": 5.0728,
393
- "step": 64000
394
- },
395
- {
396
- "epoch": 50.78,
397
- "learning_rate": 1.555555555555556e-05,
398
- "loss": 5.0692,
399
- "step": 65000
400
- },
401
- {
402
- "epoch": 51.56,
403
- "learning_rate": 1.5111111111111112e-05,
404
- "loss": 5.065,
405
- "step": 66000
406
- },
407
- {
408
- "epoch": 52.34,
409
- "learning_rate": 1.4666666666666666e-05,
410
- "loss": 5.0594,
411
- "step": 67000
412
- },
413
- {
414
- "epoch": 53.12,
415
- "learning_rate": 1.4222222222222224e-05,
416
- "loss": 5.0549,
417
- "step": 68000
418
- },
419
- {
420
- "epoch": 53.91,
421
- "learning_rate": 1.377777777777778e-05,
422
- "loss": 5.0538,
423
- "step": 69000
424
- },
425
- {
426
- "epoch": 54.69,
427
- "learning_rate": 1.3333333333333333e-05,
428
- "loss": 5.0482,
429
- "step": 70000
430
- },
431
- {
432
- "epoch": 55.47,
433
- "learning_rate": 1.288888888888889e-05,
434
- "loss": 5.0475,
435
- "step": 71000
436
- },
437
- {
438
- "epoch": 56.25,
439
- "learning_rate": 1.2444444444444446e-05,
440
- "loss": 5.0399,
441
- "step": 72000
442
- },
443
- {
444
- "epoch": 57.03,
445
- "learning_rate": 1.2e-05,
446
- "loss": 5.0406,
447
- "step": 73000
448
- },
449
- {
450
- "epoch": 57.81,
451
- "learning_rate": 1.1555555555555556e-05,
452
- "loss": 5.037,
453
- "step": 74000
454
- },
455
- {
456
- "epoch": 58.59,
457
- "learning_rate": 1.1111111111111113e-05,
458
- "loss": 5.0326,
459
- "step": 75000
460
- },
461
- {
462
- "epoch": 59.38,
463
- "learning_rate": 1.0666666666666667e-05,
464
- "loss": 5.031,
465
- "step": 76000
466
- },
467
- {
468
- "epoch": 60.16,
469
- "learning_rate": 1.0222222222222223e-05,
470
- "loss": 5.0291,
471
- "step": 77000
472
- },
473
- {
474
- "epoch": 60.94,
475
- "learning_rate": 9.777777777777779e-06,
476
- "loss": 5.0266,
477
- "step": 78000
478
- },
479
- {
480
- "epoch": 61.72,
481
- "learning_rate": 9.333333333333334e-06,
482
- "loss": 5.0235,
483
- "step": 79000
484
- },
485
- {
486
- "epoch": 62.5,
487
- "learning_rate": 8.888888888888888e-06,
488
- "loss": 5.0211,
489
- "step": 80000
490
- },
491
- {
492
- "epoch": 63.28,
493
- "learning_rate": 8.444444444444446e-06,
494
- "loss": 5.0214,
495
- "step": 81000
496
- },
497
- {
498
- "epoch": 64.06,
499
- "learning_rate": 8.000000000000001e-06,
500
- "loss": 5.0176,
501
- "step": 82000
502
- },
503
- {
504
- "epoch": 64.84,
505
- "learning_rate": 7.555555555555556e-06,
506
- "loss": 5.0157,
507
- "step": 83000
508
- },
509
- {
510
- "epoch": 65.62,
511
- "learning_rate": 7.111111111111112e-06,
512
- "loss": 5.014,
513
- "step": 84000
514
- },
515
- {
516
- "epoch": 66.41,
517
- "learning_rate": 6.666666666666667e-06,
518
- "loss": 5.0115,
519
- "step": 85000
520
- },
521
- {
522
- "epoch": 67.19,
523
- "learning_rate": 6.222222222222223e-06,
524
- "loss": 5.0087,
525
- "step": 86000
526
- },
527
- {
528
- "epoch": 67.97,
529
- "learning_rate": 5.777777777777778e-06,
530
- "loss": 5.0105,
531
- "step": 87000
532
- },
533
- {
534
- "epoch": 68.75,
535
- "learning_rate": 5.333333333333334e-06,
536
- "loss": 5.0062,
537
- "step": 88000
538
- },
539
- {
540
- "epoch": 69.53,
541
- "learning_rate": 4.888888888888889e-06,
542
- "loss": 5.0028,
543
- "step": 89000
544
- },
545
- {
546
- "epoch": 70.31,
547
- "learning_rate": 4.444444444444444e-06,
548
- "loss": 5.0066,
549
- "step": 90000
550
- },
551
- {
552
- "epoch": 71.09,
553
- "learning_rate": 4.000000000000001e-06,
554
- "loss": 5.0032,
555
- "step": 91000
556
- },
557
- {
558
- "epoch": 71.88,
559
- "learning_rate": 3.555555555555556e-06,
560
- "loss": 5.0022,
561
- "step": 92000
562
- },
563
- {
564
- "epoch": 72.66,
565
- "learning_rate": 3.1111111111111116e-06,
566
- "loss": 5.0024,
567
- "step": 93000
568
- },
569
- {
570
- "epoch": 73.44,
571
- "learning_rate": 2.666666666666667e-06,
572
- "loss": 5.0007,
573
- "step": 94000
574
- },
575
- {
576
- "epoch": 74.22,
577
- "learning_rate": 2.222222222222222e-06,
578
- "loss": 5.0008,
579
- "step": 95000
580
- },
581
- {
582
- "epoch": 75.0,
583
- "learning_rate": 1.777777777777778e-06,
584
- "loss": 4.9978,
585
- "step": 96000
586
- },
587
- {
588
- "epoch": 75.78,
589
- "learning_rate": 1.3333333333333334e-06,
590
- "loss": 4.9966,
591
- "step": 97000
592
- },
593
- {
594
- "epoch": 76.56,
595
- "learning_rate": 8.88888888888889e-07,
596
- "loss": 4.9994,
597
- "step": 98000
598
- },
599
- {
600
- "epoch": 77.34,
601
- "learning_rate": 4.444444444444445e-07,
602
- "loss": 4.9966,
603
- "step": 99000
604
- },
605
- {
606
- "epoch": 78.12,
607
- "learning_rate": 0.0,
608
- "loss": 4.9976,
609
- "step": 100000
610
- },
611
- {
612
- "epoch": 9.87,
613
- "learning_rate": 2.0842105263157895e-05,
614
- "loss": 5.0155,
615
- "step": 101000
616
- },
617
- {
618
- "epoch": 9.97,
619
- "learning_rate": 2.0631578947368425e-05,
620
- "loss": 5.024,
621
- "step": 102000
622
- },
623
- {
624
- "epoch": 10.06,
625
- "learning_rate": 2.0421052631578952e-05,
626
- "loss": 5.0241,
627
- "step": 103000
628
- },
629
- {
630
- "epoch": 10.16,
631
- "learning_rate": 2.0210526315789475e-05,
632
- "loss": 5.0218,
633
- "step": 104000
634
- },
635
- {
636
- "epoch": 10.26,
637
- "learning_rate": 2e-05,
638
- "loss": 5.0231,
639
- "step": 105000
640
- },
641
- {
642
- "epoch": 10.36,
643
- "learning_rate": 1.9789473684210528e-05,
644
- "loss": 5.0231,
645
- "step": 106000
646
- },
647
- {
648
- "epoch": 10.45,
649
- "learning_rate": 1.9578947368421055e-05,
650
- "loss": 5.0198,
651
- "step": 107000
652
- },
653
- {
654
- "epoch": 10.55,
655
- "learning_rate": 1.936842105263158e-05,
656
- "loss": 5.0158,
657
- "step": 108000
658
- },
659
- {
660
- "epoch": 10.65,
661
- "learning_rate": 1.9157894736842108e-05,
662
- "loss": 5.0122,
663
- "step": 109000
664
- },
665
- {
666
- "epoch": 10.75,
667
- "learning_rate": 1.894736842105263e-05,
668
- "loss": 5.0165,
669
- "step": 110000
670
- },
671
- {
672
- "epoch": 10.85,
673
- "learning_rate": 1.873684210526316e-05,
674
- "loss": 5.0183,
675
- "step": 111000
676
- },
677
- {
678
- "epoch": 10.94,
679
- "learning_rate": 1.8526315789473684e-05,
680
- "loss": 5.01,
681
- "step": 112000
682
- },
683
- {
684
- "epoch": 11.04,
685
- "learning_rate": 1.831578947368421e-05,
686
- "loss": 5.007,
687
- "step": 113000
688
- },
689
- {
690
- "epoch": 11.14,
691
- "learning_rate": 1.810526315789474e-05,
692
- "loss": 5.0059,
693
- "step": 114000
694
- },
695
- {
696
- "epoch": 11.24,
697
- "learning_rate": 1.7894736842105264e-05,
698
- "loss": 5.0089,
699
- "step": 115000
700
- },
701
- {
702
- "epoch": 11.33,
703
- "learning_rate": 1.768421052631579e-05,
704
- "loss": 5.0081,
705
- "step": 116000
706
- },
707
- {
708
- "epoch": 11.43,
709
- "learning_rate": 1.7473684210526317e-05,
710
- "loss": 5.002,
711
- "step": 117000
712
- },
713
- {
714
- "epoch": 11.53,
715
- "learning_rate": 1.7263157894736843e-05,
716
- "loss": 5.0052,
717
- "step": 118000
718
- },
719
- {
720
- "epoch": 11.63,
721
- "learning_rate": 1.705263157894737e-05,
722
- "loss": 5.0079,
723
- "step": 119000
724
- },
725
- {
726
- "epoch": 11.72,
727
- "learning_rate": 1.6842105263157896e-05,
728
- "loss": 4.9985,
729
- "step": 120000
730
- },
731
- {
732
- "epoch": 11.82,
733
- "learning_rate": 1.6631578947368423e-05,
734
- "loss": 5.006,
735
- "step": 121000
736
- },
737
- {
738
- "epoch": 11.92,
739
- "learning_rate": 1.642105263157895e-05,
740
- "loss": 4.9956,
741
- "step": 122000
742
- },
743
- {
744
- "epoch": 12.02,
745
- "learning_rate": 1.6210526315789473e-05,
746
- "loss": 4.9973,
747
- "step": 123000
748
- },
749
- {
750
- "epoch": 12.12,
751
- "learning_rate": 1.6000000000000003e-05,
752
- "loss": 4.9976,
753
- "step": 124000
754
- },
755
- {
756
- "epoch": 12.21,
757
- "learning_rate": 1.578947368421053e-05,
758
- "loss": 4.9882,
759
- "step": 125000
760
- },
761
- {
762
- "epoch": 12.31,
763
- "learning_rate": 1.5578947368421052e-05,
764
- "loss": 4.9908,
765
- "step": 126000
766
- },
767
- {
768
- "epoch": 12.41,
769
- "learning_rate": 1.536842105263158e-05,
770
- "loss": 5.0038,
771
- "step": 127000
772
- },
773
- {
774
- "epoch": 12.51,
775
- "learning_rate": 1.5157894736842107e-05,
776
- "loss": 4.9923,
777
- "step": 128000
778
- },
779
- {
780
- "epoch": 12.6,
781
- "learning_rate": 1.4947368421052632e-05,
782
- "loss": 4.9847,
783
- "step": 129000
784
- },
785
- {
786
- "epoch": 12.7,
787
- "learning_rate": 1.4736842105263159e-05,
788
- "loss": 4.9864,
789
- "step": 130000
790
- },
791
- {
792
- "epoch": 12.8,
793
- "learning_rate": 1.4526315789473687e-05,
794
- "loss": 4.9903,
795
- "step": 131000
796
- },
797
- {
798
- "epoch": 12.9,
799
- "learning_rate": 1.4315789473684212e-05,
800
- "loss": 4.9812,
801
- "step": 132000
802
- },
803
- {
804
- "epoch": 12.99,
805
- "learning_rate": 1.4105263157894738e-05,
806
- "loss": 4.9923,
807
- "step": 133000
808
- },
809
- {
810
- "epoch": 13.09,
811
- "learning_rate": 1.3894736842105265e-05,
812
- "loss": 4.9838,
813
- "step": 134000
814
- },
815
- {
816
- "epoch": 13.19,
817
- "learning_rate": 1.3684210526315791e-05,
818
- "loss": 4.977,
819
- "step": 135000
820
- },
821
- {
822
- "epoch": 13.29,
823
- "learning_rate": 1.3473684210526316e-05,
824
- "loss": 4.9743,
825
- "step": 136000
826
- },
827
- {
828
- "epoch": 13.39,
829
- "learning_rate": 1.3263157894736843e-05,
830
- "loss": 4.9797,
831
- "step": 137000
832
- },
833
- {
834
- "epoch": 13.48,
835
- "learning_rate": 1.305263157894737e-05,
836
- "loss": 4.9915,
837
- "step": 138000
838
- },
839
- {
840
- "epoch": 13.58,
841
- "learning_rate": 1.2842105263157896e-05,
842
- "loss": 4.9762,
843
- "step": 139000
844
- },
845
- {
846
- "epoch": 13.68,
847
- "learning_rate": 1.263157894736842e-05,
848
- "loss": 4.9771,
849
- "step": 140000
850
- },
851
- {
852
- "epoch": 13.78,
853
- "learning_rate": 1.2421052631578949e-05,
854
- "loss": 4.9763,
855
- "step": 141000
856
- },
857
- {
858
- "epoch": 13.87,
859
- "learning_rate": 1.2210526315789475e-05,
860
- "loss": 4.9706,
861
- "step": 142000
862
- },
863
- {
864
- "epoch": 13.97,
865
- "learning_rate": 1.2e-05,
866
- "loss": 4.9731,
867
- "step": 143000
868
- },
869
- {
870
- "epoch": 14.07,
871
- "learning_rate": 1.1789473684210527e-05,
872
- "loss": 4.9736,
873
- "step": 144000
874
- },
875
- {
876
- "epoch": 14.17,
877
- "learning_rate": 1.1578947368421053e-05,
878
- "loss": 4.9704,
879
- "step": 145000
880
- },
881
- {
882
- "epoch": 14.26,
883
- "learning_rate": 1.136842105263158e-05,
884
- "loss": 4.9762,
885
- "step": 146000
886
- },
887
- {
888
- "epoch": 14.36,
889
- "learning_rate": 1.1157894736842105e-05,
890
- "loss": 4.9709,
891
- "step": 147000
892
- },
893
- {
894
- "epoch": 14.46,
895
- "learning_rate": 1.0947368421052633e-05,
896
- "loss": 4.957,
897
- "step": 148000
898
- },
899
- {
900
- "epoch": 14.56,
901
- "learning_rate": 1.073684210526316e-05,
902
- "loss": 4.9666,
903
- "step": 149000
904
- },
905
- {
906
- "epoch": 14.66,
907
- "learning_rate": 1.0526315789473684e-05,
908
- "loss": 4.9623,
909
- "step": 150000
910
- },
911
- {
912
- "epoch": 14.75,
913
- "learning_rate": 1.0315789473684213e-05,
914
- "loss": 4.9613,
915
- "step": 151000
916
- },
917
- {
918
- "epoch": 14.85,
919
- "learning_rate": 1.0105263157894738e-05,
920
- "loss": 4.9642,
921
- "step": 152000
922
- },
923
- {
924
- "epoch": 14.95,
925
- "learning_rate": 9.894736842105264e-06,
926
- "loss": 4.9605,
927
- "step": 153000
928
- },
929
- {
930
- "epoch": 15.05,
931
- "learning_rate": 9.68421052631579e-06,
932
- "loss": 4.9688,
933
- "step": 154000
934
- },
935
- {
936
- "epoch": 15.14,
937
- "learning_rate": 9.473684210526315e-06,
938
- "loss": 4.9635,
939
- "step": 155000
940
- },
941
- {
942
- "epoch": 15.24,
943
- "learning_rate": 9.263157894736842e-06,
944
- "loss": 4.9587,
945
- "step": 156000
946
- },
947
- {
948
- "epoch": 15.34,
949
- "learning_rate": 9.05263157894737e-06,
950
- "loss": 4.9633,
951
- "step": 157000
952
- },
953
- {
954
- "epoch": 15.44,
955
- "learning_rate": 8.842105263157895e-06,
956
- "loss": 4.9613,
957
- "step": 158000
958
- },
959
- {
960
- "epoch": 15.53,
961
- "learning_rate": 8.631578947368422e-06,
962
- "loss": 4.9629,
963
- "step": 159000
964
- },
965
- {
966
- "epoch": 15.63,
967
- "learning_rate": 8.421052631578948e-06,
968
- "loss": 4.9547,
969
- "step": 160000
970
- },
971
- {
972
- "epoch": 15.73,
973
- "learning_rate": 8.210526315789475e-06,
974
- "loss": 4.9541,
975
- "step": 161000
976
- },
977
- {
978
- "epoch": 15.83,
979
- "learning_rate": 8.000000000000001e-06,
980
- "loss": 4.9532,
981
- "step": 162000
982
- },
983
- {
984
- "epoch": 15.93,
985
- "learning_rate": 7.789473684210526e-06,
986
- "loss": 4.9461,
987
- "step": 163000
988
- },
989
- {
990
- "epoch": 16.02,
991
- "learning_rate": 7.578947368421054e-06,
992
- "loss": 4.9518,
993
- "step": 164000
994
- },
995
- {
996
- "epoch": 16.12,
997
- "learning_rate": 7.368421052631579e-06,
998
- "loss": 4.9543,
999
- "step": 165000
1000
- },
1001
- {
1002
- "epoch": 16.22,
1003
- "learning_rate": 7.157894736842106e-06,
1004
- "loss": 4.9526,
1005
- "step": 166000
1006
- },
1007
- {
1008
- "epoch": 16.32,
1009
- "learning_rate": 6.947368421052632e-06,
1010
- "loss": 4.9447,
1011
- "step": 167000
1012
- },
1013
- {
1014
- "epoch": 16.41,
1015
- "learning_rate": 6.736842105263158e-06,
1016
- "loss": 4.9545,
1017
- "step": 168000
1018
- },
1019
- {
1020
- "epoch": 16.51,
1021
- "learning_rate": 6.526315789473685e-06,
1022
- "loss": 4.9448,
1023
- "step": 169000
1024
- },
1025
- {
1026
- "epoch": 16.61,
1027
- "learning_rate": 6.31578947368421e-06,
1028
- "loss": 4.9521,
1029
- "step": 170000
1030
- },
1031
- {
1032
- "epoch": 16.71,
1033
- "learning_rate": 6.105263157894738e-06,
1034
- "loss": 4.9503,
1035
- "step": 171000
1036
- },
1037
- {
1038
- "epoch": 16.81,
1039
- "learning_rate": 5.8947368421052634e-06,
1040
- "loss": 4.9537,
1041
- "step": 172000
1042
- },
1043
- {
1044
- "epoch": 16.9,
1045
- "learning_rate": 5.68421052631579e-06,
1046
- "loss": 4.9446,
1047
- "step": 173000
1048
- },
1049
- {
1050
- "epoch": 17.0,
1051
- "learning_rate": 5.4736842105263165e-06,
1052
- "loss": 4.9397,
1053
- "step": 174000
1054
- },
1055
- {
1056
- "epoch": 17.1,
1057
- "learning_rate": 5.263157894736842e-06,
1058
- "loss": 4.9541,
1059
- "step": 175000
1060
- },
1061
- {
1062
- "epoch": 17.2,
1063
- "learning_rate": 5.052631578947369e-06,
1064
- "loss": 4.9449,
1065
- "step": 176000
1066
- },
1067
- {
1068
- "epoch": 17.29,
1069
- "learning_rate": 4.842105263157895e-06,
1070
- "loss": 4.9447,
1071
- "step": 177000
1072
- },
1073
- {
1074
- "epoch": 17.39,
1075
- "learning_rate": 4.631578947368421e-06,
1076
- "loss": 4.9344,
1077
- "step": 178000
1078
- },
1079
- {
1080
- "epoch": 17.49,
1081
- "learning_rate": 4.4210526315789476e-06,
1082
- "loss": 4.933,
1083
- "step": 179000
1084
- },
1085
- {
1086
- "epoch": 17.59,
1087
- "learning_rate": 4.210526315789474e-06,
1088
- "loss": 4.9397,
1089
- "step": 180000
1090
- },
1091
- {
1092
- "epoch": 17.68,
1093
- "learning_rate": 4.000000000000001e-06,
1094
- "loss": 4.9431,
1095
- "step": 181000
1096
- },
1097
- {
1098
- "epoch": 17.78,
1099
- "learning_rate": 3.789473684210527e-06,
1100
- "loss": 4.9378,
1101
- "step": 182000
1102
- },
1103
- {
1104
- "epoch": 17.88,
1105
- "learning_rate": 3.578947368421053e-06,
1106
- "loss": 4.9399,
1107
- "step": 183000
1108
- },
1109
- {
1110
- "epoch": 17.98,
1111
- "learning_rate": 3.368421052631579e-06,
1112
- "loss": 4.9481,
1113
- "step": 184000
1114
- },
1115
- {
1116
- "epoch": 18.08,
1117
- "learning_rate": 3.157894736842105e-06,
1118
- "loss": 4.9399,
1119
- "step": 185000
1120
- },
1121
- {
1122
- "epoch": 18.17,
1123
- "learning_rate": 2.9473684210526317e-06,
1124
- "loss": 4.9365,
1125
- "step": 186000
1126
- },
1127
- {
1128
- "epoch": 18.27,
1129
- "learning_rate": 2.7368421052631583e-06,
1130
- "loss": 4.9307,
1131
- "step": 187000
1132
- },
1133
- {
1134
- "epoch": 18.37,
1135
- "learning_rate": 2.5263157894736844e-06,
1136
- "loss": 4.9391,
1137
- "step": 188000
1138
- },
1139
- {
1140
- "epoch": 18.47,
1141
- "learning_rate": 2.3157894736842105e-06,
1142
- "loss": 4.9331,
1143
- "step": 189000
1144
- },
1145
- {
1146
- "epoch": 18.56,
1147
- "learning_rate": 2.105263157894737e-06,
1148
- "loss": 4.948,
1149
- "step": 190000
1150
- },
1151
- {
1152
- "epoch": 18.66,
1153
- "learning_rate": 1.8947368421052634e-06,
1154
- "loss": 4.9407,
1155
- "step": 191000
1156
- },
1157
- {
1158
- "epoch": 18.76,
1159
- "learning_rate": 1.6842105263157895e-06,
1160
- "loss": 4.944,
1161
- "step": 192000
1162
- },
1163
- {
1164
- "epoch": 18.86,
1165
- "learning_rate": 1.4736842105263159e-06,
1166
- "loss": 4.9286,
1167
- "step": 193000
1168
- },
1169
- {
1170
- "epoch": 18.95,
1171
- "learning_rate": 1.2631578947368422e-06,
1172
- "loss": 4.9338,
1173
- "step": 194000
1174
- },
1175
- {
1176
- "epoch": 19.05,
1177
- "learning_rate": 1.0526315789473685e-06,
1178
- "loss": 4.9357,
1179
- "step": 195000
1180
- }
1181
- ],
1182
- "logging_steps": 1000,
1183
- "max_steps": 200000,
1184
- "num_train_epochs": 20,
1185
- "save_steps": 1000,
1186
- "total_flos": 2.2607703469056328e+19,
1187
- "trial_name": null,
1188
- "trial_params": null
1189
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/checkpoint-195000/training_args.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:c9ff5d6388a59708cb6305a7fab984d438e385b1835680f9629b90c23c5cf704
3
- size 5137
 
 
 
 
len2_5120/checkpoint-196000/config.json DELETED
@@ -1,25 +0,0 @@
1
- {
2
- "architectures": [
3
- "BertForMaskedLM"
4
- ],
5
- "attention_probs_dropout_prob": 0.1,
6
- "classifier_dropout": null,
7
- "hidden_act": "gelu",
8
- "hidden_dropout_prob": 0.1,
9
- "hidden_size": 768,
10
- "initializer_range": 0.02,
11
- "intermediate_size": 3072,
12
- "layer_norm_eps": 1e-12,
13
- "max_length": 512,
14
- "max_position_embeddings": 512,
15
- "model_type": "bert",
16
- "num_attention_heads": 12,
17
- "num_hidden_layers": 12,
18
- "pad_token_id": 0,
19
- "position_embedding_type": "absolute",
20
- "torch_dtype": "float32",
21
- "transformers_version": "4.35.2",
22
- "type_vocab_size": 2,
23
- "use_cache": true,
24
- "vocab_size": 5120
25
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
len2_5120/checkpoint-196000/generation_config.json DELETED
@@ -1,6 +0,0 @@
1
- {
2
- "_from_model_config": true,
3
- "max_length": 512,
4
- "pad_token_id": 0,
5
- "transformers_version": "4.35.2"
6
- }
 
 
 
 
 
 
 
len2_5120/checkpoint-196000/model.safetensors DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:93b5c27f5deb7b1b609b2d5cc9ea6c0929507d7419b4a7b7c5c13656d66d9b0f
3
- size 359944256
 
 
 
 
len2_5120/checkpoint-196000/optimizer.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:73372e579073fdfb073e755d42fa884d0e55d89671ba5e0fac500afa0ac9c72b
3
- size 720010699
 
 
 
 
len2_5120/checkpoint-196000/rng_state.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:fafaef33931b73eb438d9dcbd75cbb6a779009b5570372ee03872e55b8f5690c
3
- size 14645