Hugging Face's logo

nlphuji
/
mt_coref_en_ru_coref_exp

Text2Text Generation
Transformers
PyTorch
English
Russian
marian
Generated from Trainer
Inference Endpoints
Model card Files Community
1
cattana commited on
Commit
a6b41f7
•
1 Parent(s): a136d77

add missing files

Browse files
Files changed (14) hide show
  1. .gitattributes +1 -0
  2. README.md +54 -0
  3. added_tokens.json +1 -0
  4. all_results.json +8 -0
  5. config.json +2 -3
  6. pytorch_model.bin +2 -2
  7. source.spm +0 -0
  8. special_tokens_map.json +1 -0
  9. target.spm +3 -0
  10. tokenizer_config.json +1 -0
  11. train_results.json +8 -0
  12. trainer_state.json +2815 -0
  13. training_args.bin +3 -0
  14. vocab.json +0 -0
.gitattributes CHANGED
@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
32
  *.zip filter=lfs diff=lfs merge=lfs -text
33
  *.zst filter=lfs diff=lfs merge=lfs -text
34
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
32
  *.zip filter=lfs diff=lfs merge=lfs -text
33
  *.zst filter=lfs diff=lfs merge=lfs -text
34
  *tfevents* filter=lfs diff=lfs merge=lfs -text
35
+ target.spm filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ - ru
5
+ license: apache-2.0
6
+ tags:
7
+ - generated_from_trainer
8
+ model-index:
9
+ - name: ru_coref_all
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # ru_coref_all
17
+
18
+ This model is a fine-tuned version of [Helsinki-NLP/opus-mt-en-ru](https://huggingface.co/Helsinki-NLP/opus-mt-en-ru) on an unknown dataset.
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
+
32
+ ## Training procedure
33
+
34
+ ### Training hyperparameters
35
+
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 5e-05
38
+ - train_batch_size: 16
39
+ - eval_batch_size: 16
40
+ - seed: 42
41
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
+ - lr_scheduler_type: linear
43
+ - num_epochs: 3.0
44
+
45
+ ### Training results
46
+
47
+
48
+
49
+ ### Framework versions
50
+
51
+ - Transformers 4.16.0.dev0
52
+ - Pytorch 1.10.1+cu102
53
+ - Datasets 1.17.0
54
+ - Tokenizers 0.10.3
added_tokens.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"<ENT0>": 62518, "</ENT0>": 62519, "<ENT1>": 62520, "</ENT1>": 62521, "<ENT2>": 62522, "</ENT2>": 62523, "<ENT3>": 62524, "</ENT3>": 62525, "<ENT4>": 62526, "</ENT4>": 62527, "<ENT5>": 62528, "</ENT5>": 62529, "<ENT6>": 62530, "</ENT6>": 62531, "<ENT7>": 62532, "</ENT7>": 62533, "<ENT8>": 62534, "</ENT8>": 62535, "<ENT9>": 62536, "</ENT9>": 62537, "<ENT10>": 62538, "</ENT10>": 62539, "<ENT11>": 62540, "</ENT11>": 62541, "<ENT12>": 62542, "</ENT12>": 62543, "<ENT13>": 62544, "</ENT13>": 62545, "<ENT14>": 62546, "</ENT14>": 62547, "<ENT15>": 62548, "</ENT15>": 62549, "<ENT16>": 62550, "</ENT16>": 62551, "<ENT17>": 62552, "</ENT17>": 62553, "<ENT18>": 62554, "</ENT18>": 62555, "<ENT19>": 62556, "</ENT19>": 62557, "<MALE>": 62558, "</MALE>": 62559, "<FEMALE>": 62560, "</FEMALE>": 62561}
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 1.506600116780431,
4
+ "train_runtime": 70430.9377,
5
+ "train_samples": 1240884,
6
+ "train_samples_per_second": 52.855,
7
+ "train_steps_per_second": 3.303
8
+ }
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "/cs/labs/oabend/asaf305612939/co-ref-project/mt-output/paper_models/ru/en_ru_coref_exp",
3
  "_num_labels": 3,
4
  "activation_dropout": 0.0,
5
  "activation_function": "swish",
@@ -30,7 +30,6 @@
30
  "encoder_layers": 6,
31
  "eos_token_id": 0,
32
  "forced_eos_token_id": 0,
33
- "gradient_checkpointing": false,
34
  "id2label": {
35
  "0": "LABEL_0",
36
  "1": "LABEL_1",
@@ -54,7 +53,7 @@
54
  "scale_embedding": true,
55
  "static_position_embeddings": true,
56
  "torch_dtype": "float32",
57
- "transformers_version": "4.9.0.dev0",
58
  "use_cache": true,
59
  "vocab_size": 62562
60
  }
 
1
  {
2
+ "_name_or_path": "Helsinki-NLP/opus-mt-en-ru",
3
  "_num_labels": 3,
4
  "activation_dropout": 0.0,
5
  "activation_function": "swish",
 
30
  "encoder_layers": 6,
31
  "eos_token_id": 0,
32
  "forced_eos_token_id": 0,
 
33
  "id2label": {
34
  "0": "LABEL_0",
35
  "1": "LABEL_1",
 
53
  "scale_embedding": true,
54
  "static_position_embeddings": true,
55
  "torch_dtype": "float32",
56
+ "transformers_version": "4.16.0.dev0",
57
  "use_cache": true,
58
  "vocab_size": 62562
59
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5e9085fa07623298772603e5e4455292360336dd51df37c2bc816077b5a66d3e
3
- size 305018461
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ab5155127585fe32254be12b525958f5cb4ef2cc1f459dcfb8ddb83bb96a935
3
+ size 305025605
source.spm ADDED
Binary file (803 kB). View file
 
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
target.spm ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:745998e51ba5b058e38b7ac7765c25c43ed5c1c39cc92b27163b9b2e323c9d7c
3
+ size 1080169
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"source_lang": "en", "target_lang": "ru", "unk_token": "<unk>", "eos_token": "</s>", "pad_token": "<pad>", "model_max_length": 512, "sp_model_kwargs": {}, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "Helsinki-NLP/opus-mt-en-ru", "tokenizer_class": "MarianTokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 1.506600116780431,
4
+ "train_runtime": 70430.9377,
5
+ "train_samples": 1240884,
6
+ "train_samples_per_second": 52.855,
7
+ "train_steps_per_second": 3.303
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,2815 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "global_step": 232668,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 4.989255075902144e-05,
13
+ "loss": 1.8394,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 4.978510151804288e-05,
19
+ "loss": 1.8157,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.02,
24
+ "learning_rate": 4.967765227706432e-05,
25
+ "loss": 1.7863,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.03,
30
+ "learning_rate": 4.9570203036085755e-05,
31
+ "loss": 1.7806,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.03,
36
+ "learning_rate": 4.946275379510719e-05,
37
+ "loss": 1.7832,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.04,
42
+ "learning_rate": 4.935530455412863e-05,
43
+ "loss": 1.7694,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.05,
48
+ "learning_rate": 4.924785531315007e-05,
49
+ "loss": 1.7631,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.05,
54
+ "learning_rate": 4.914040607217151e-05,
55
+ "loss": 1.7652,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.06,
60
+ "learning_rate": 4.9032956831192946e-05,
61
+ "loss": 1.7578,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.06,
66
+ "learning_rate": 4.8925507590214384e-05,
67
+ "loss": 1.7579,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.07,
72
+ "learning_rate": 4.881805834923582e-05,
73
+ "loss": 1.7552,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.08,
78
+ "learning_rate": 4.871060910825726e-05,
79
+ "loss": 1.7485,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.08,
84
+ "learning_rate": 4.8603159867278705e-05,
85
+ "loss": 1.7474,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.09,
90
+ "learning_rate": 4.849571062630014e-05,
91
+ "loss": 1.7557,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.1,
96
+ "learning_rate": 4.8388261385321575e-05,
97
+ "loss": 1.7486,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.1,
102
+ "learning_rate": 4.828081214434301e-05,
103
+ "loss": 1.7357,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.11,
108
+ "learning_rate": 4.817336290336446e-05,
109
+ "loss": 1.7341,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.12,
114
+ "learning_rate": 4.806591366238589e-05,
115
+ "loss": 1.7298,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.12,
120
+ "learning_rate": 4.795846442140733e-05,
121
+ "loss": 1.7354,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 0.13,
126
+ "learning_rate": 4.7851015180428766e-05,
127
+ "loss": 1.7269,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 0.14,
132
+ "learning_rate": 4.774356593945021e-05,
133
+ "loss": 1.7171,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 0.14,
138
+ "learning_rate": 4.763611669847164e-05,
139
+ "loss": 1.715,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 0.15,
144
+ "learning_rate": 4.752866745749308e-05,
145
+ "loss": 1.7232,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 0.15,
150
+ "learning_rate": 4.7421218216514525e-05,
151
+ "loss": 1.7249,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 0.16,
156
+ "learning_rate": 4.731376897553596e-05,
157
+ "loss": 1.7187,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 0.17,
162
+ "learning_rate": 4.7206319734557395e-05,
163
+ "loss": 1.7095,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 0.17,
168
+ "learning_rate": 4.709887049357883e-05,
169
+ "loss": 1.7193,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 0.18,
174
+ "learning_rate": 4.699142125260028e-05,
175
+ "loss": 1.7075,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 0.19,
180
+ "learning_rate": 4.6883972011621716e-05,
181
+ "loss": 1.7069,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 0.19,
186
+ "learning_rate": 4.677652277064315e-05,
187
+ "loss": 1.7119,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 0.2,
192
+ "learning_rate": 4.6669073529664586e-05,
193
+ "loss": 1.6998,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 0.21,
198
+ "learning_rate": 4.656162428868603e-05,
199
+ "loss": 1.7022,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 0.21,
204
+ "learning_rate": 4.645417504770747e-05,
205
+ "loss": 1.6983,
206
+ "step": 16500
207
+ },
208
+ {
209
+ "epoch": 0.22,
210
+ "learning_rate": 4.63467258067289e-05,
211
+ "loss": 1.6967,
212
+ "step": 17000
213
+ },
214
+ {
215
+ "epoch": 0.23,
216
+ "learning_rate": 4.623927656575034e-05,
217
+ "loss": 1.6953,
218
+ "step": 17500
219
+ },
220
+ {
221
+ "epoch": 0.23,
222
+ "learning_rate": 4.613182732477178e-05,
223
+ "loss": 1.7046,
224
+ "step": 18000
225
+ },
226
+ {
227
+ "epoch": 0.24,
228
+ "learning_rate": 4.602437808379322e-05,
229
+ "loss": 1.7041,
230
+ "step": 18500
231
+ },
232
+ {
233
+ "epoch": 0.24,
234
+ "learning_rate": 4.591692884281465e-05,
235
+ "loss": 1.6965,
236
+ "step": 19000
237
+ },
238
+ {
239
+ "epoch": 0.25,
240
+ "learning_rate": 4.58094796018361e-05,
241
+ "loss": 1.7069,
242
+ "step": 19500
243
+ },
244
+ {
245
+ "epoch": 0.26,
246
+ "learning_rate": 4.5702030360857536e-05,
247
+ "loss": 1.6918,
248
+ "step": 20000
249
+ },
250
+ {
251
+ "epoch": 0.26,
252
+ "learning_rate": 4.5594581119878974e-05,
253
+ "loss": 1.6731,
254
+ "step": 20500
255
+ },
256
+ {
257
+ "epoch": 0.27,
258
+ "learning_rate": 4.5487131878900405e-05,
259
+ "loss": 1.6889,
260
+ "step": 21000
261
+ },
262
+ {
263
+ "epoch": 0.28,
264
+ "learning_rate": 4.537968263792185e-05,
265
+ "loss": 1.684,
266
+ "step": 21500
267
+ },
268
+ {
269
+ "epoch": 0.28,
270
+ "learning_rate": 4.527223339694329e-05,
271
+ "loss": 1.7031,
272
+ "step": 22000
273
+ },
274
+ {
275
+ "epoch": 0.29,
276
+ "learning_rate": 4.516478415596473e-05,
277
+ "loss": 1.6796,
278
+ "step": 22500
279
+ },
280
+ {
281
+ "epoch": 0.3,
282
+ "learning_rate": 4.505733491498616e-05,
283
+ "loss": 1.6741,
284
+ "step": 23000
285
+ },
286
+ {
287
+ "epoch": 0.3,
288
+ "learning_rate": 4.49498856740076e-05,
289
+ "loss": 1.6672,
290
+ "step": 23500
291
+ },
292
+ {
293
+ "epoch": 0.31,
294
+ "learning_rate": 4.484243643302904e-05,
295
+ "loss": 1.672,
296
+ "step": 24000
297
+ },
298
+ {
299
+ "epoch": 0.32,
300
+ "learning_rate": 4.473498719205047e-05,
301
+ "loss": 1.6845,
302
+ "step": 24500
303
+ },
304
+ {
305
+ "epoch": 0.32,
306
+ "learning_rate": 4.462753795107192e-05,
307
+ "loss": 1.6756,
308
+ "step": 25000
309
+ },
310
+ {
311
+ "epoch": 0.33,
312
+ "learning_rate": 4.4520088710093356e-05,
313
+ "loss": 1.6776,
314
+ "step": 25500
315
+ },
316
+ {
317
+ "epoch": 0.34,
318
+ "learning_rate": 4.4412639469114794e-05,
319
+ "loss": 1.6742,
320
+ "step": 26000
321
+ },
322
+ {
323
+ "epoch": 0.34,
324
+ "learning_rate": 4.4305190228136225e-05,
325
+ "loss": 1.7051,
326
+ "step": 26500
327
+ },
328
+ {
329
+ "epoch": 0.35,
330
+ "learning_rate": 4.419774098715767e-05,
331
+ "loss": 1.6833,
332
+ "step": 27000
333
+ },
334
+ {
335
+ "epoch": 0.35,
336
+ "learning_rate": 4.409029174617911e-05,
337
+ "loss": 1.6652,
338
+ "step": 27500
339
+ },
340
+ {
341
+ "epoch": 0.36,
342
+ "learning_rate": 4.3982842505200547e-05,
343
+ "loss": 1.6562,
344
+ "step": 28000
345
+ },
346
+ {
347
+ "epoch": 0.37,
348
+ "learning_rate": 4.387539326422198e-05,
349
+ "loss": 1.6777,
350
+ "step": 28500
351
+ },
352
+ {
353
+ "epoch": 0.37,
354
+ "learning_rate": 4.376794402324342e-05,
355
+ "loss": 1.6822,
356
+ "step": 29000
357
+ },
358
+ {
359
+ "epoch": 0.38,
360
+ "learning_rate": 4.366049478226486e-05,
361
+ "loss": 1.6758,
362
+ "step": 29500
363
+ },
364
+ {
365
+ "epoch": 0.39,
366
+ "learning_rate": 4.35530455412863e-05,
367
+ "loss": 1.675,
368
+ "step": 30000
369
+ },
370
+ {
371
+ "epoch": 0.39,
372
+ "learning_rate": 4.344559630030774e-05,
373
+ "loss": 1.6585,
374
+ "step": 30500
375
+ },
376
+ {
377
+ "epoch": 0.4,
378
+ "learning_rate": 4.3338147059329176e-05,
379
+ "loss": 1.6532,
380
+ "step": 31000
381
+ },
382
+ {
383
+ "epoch": 0.41,
384
+ "learning_rate": 4.3230697818350614e-05,
385
+ "loss": 1.6698,
386
+ "step": 31500
387
+ },
388
+ {
389
+ "epoch": 0.41,
390
+ "learning_rate": 4.312324857737205e-05,
391
+ "loss": 1.6591,
392
+ "step": 32000
393
+ },
394
+ {
395
+ "epoch": 0.42,
396
+ "learning_rate": 4.301579933639349e-05,
397
+ "loss": 1.6563,
398
+ "step": 32500
399
+ },
400
+ {
401
+ "epoch": 0.43,
402
+ "learning_rate": 4.290835009541493e-05,
403
+ "loss": 1.6749,
404
+ "step": 33000
405
+ },
406
+ {
407
+ "epoch": 0.43,
408
+ "learning_rate": 4.2800900854436366e-05,
409
+ "loss": 1.6536,
410
+ "step": 33500
411
+ },
412
+ {
413
+ "epoch": 0.44,
414
+ "learning_rate": 4.2693451613457805e-05,
415
+ "loss": 1.6634,
416
+ "step": 34000
417
+ },
418
+ {
419
+ "epoch": 0.44,
420
+ "learning_rate": 4.258600237247924e-05,
421
+ "loss": 1.6562,
422
+ "step": 34500
423
+ },
424
+ {
425
+ "epoch": 0.45,
426
+ "learning_rate": 4.247855313150068e-05,
427
+ "loss": 1.6642,
428
+ "step": 35000
429
+ },
430
+ {
431
+ "epoch": 0.46,
432
+ "learning_rate": 4.237110389052212e-05,
433
+ "loss": 1.6493,
434
+ "step": 35500
435
+ },
436
+ {
437
+ "epoch": 0.46,
438
+ "learning_rate": 4.226365464954356e-05,
439
+ "loss": 1.6488,
440
+ "step": 36000
441
+ },
442
+ {
443
+ "epoch": 0.47,
444
+ "learning_rate": 4.2156205408564995e-05,
445
+ "loss": 1.6384,
446
+ "step": 36500
447
+ },
448
+ {
449
+ "epoch": 0.48,
450
+ "learning_rate": 4.2048756167586434e-05,
451
+ "loss": 1.6525,
452
+ "step": 37000
453
+ },
454
+ {
455
+ "epoch": 0.48,
456
+ "learning_rate": 4.194130692660787e-05,
457
+ "loss": 1.6467,
458
+ "step": 37500
459
+ },
460
+ {
461
+ "epoch": 0.49,
462
+ "learning_rate": 4.183385768562931e-05,
463
+ "loss": 1.6754,
464
+ "step": 38000
465
+ },
466
+ {
467
+ "epoch": 0.5,
468
+ "learning_rate": 4.172640844465075e-05,
469
+ "loss": 1.6594,
470
+ "step": 38500
471
+ },
472
+ {
473
+ "epoch": 0.5,
474
+ "learning_rate": 4.1618959203672186e-05,
475
+ "loss": 1.6366,
476
+ "step": 39000
477
+ },
478
+ {
479
+ "epoch": 0.51,
480
+ "learning_rate": 4.1511509962693624e-05,
481
+ "loss": 1.6476,
482
+ "step": 39500
483
+ },
484
+ {
485
+ "epoch": 0.52,
486
+ "learning_rate": 4.140406072171506e-05,
487
+ "loss": 1.6223,
488
+ "step": 40000
489
+ },
490
+ {
491
+ "epoch": 0.52,
492
+ "learning_rate": 4.12966114807365e-05,
493
+ "loss": 1.6332,
494
+ "step": 40500
495
+ },
496
+ {
497
+ "epoch": 0.53,
498
+ "learning_rate": 4.118916223975794e-05,
499
+ "loss": 1.6446,
500
+ "step": 41000
501
+ },
502
+ {
503
+ "epoch": 0.54,
504
+ "learning_rate": 4.108171299877938e-05,
505
+ "loss": 1.6336,
506
+ "step": 41500
507
+ },
508
+ {
509
+ "epoch": 0.54,
510
+ "learning_rate": 4.0974263757800815e-05,
511
+ "loss": 1.6454,
512
+ "step": 42000
513
+ },
514
+ {
515
+ "epoch": 0.55,
516
+ "learning_rate": 4.086681451682225e-05,
517
+ "loss": 1.6293,
518
+ "step": 42500
519
+ },
520
+ {
521
+ "epoch": 0.55,
522
+ "learning_rate": 4.075936527584369e-05,
523
+ "loss": 1.6485,
524
+ "step": 43000
525
+ },
526
+ {
527
+ "epoch": 0.56,
528
+ "learning_rate": 4.0651916034865136e-05,
529
+ "loss": 1.6426,
530
+ "step": 43500
531
+ },
532
+ {
533
+ "epoch": 0.57,
534
+ "learning_rate": 4.054446679388657e-05,
535
+ "loss": 1.638,
536
+ "step": 44000
537
+ },
538
+ {
539
+ "epoch": 0.57,
540
+ "learning_rate": 4.0437017552908006e-05,
541
+ "loss": 1.6456,
542
+ "step": 44500
543
+ },
544
+ {
545
+ "epoch": 0.58,
546
+ "learning_rate": 4.0329568311929444e-05,
547
+ "loss": 1.6456,
548
+ "step": 45000
549
+ },
550
+ {
551
+ "epoch": 0.59,
552
+ "learning_rate": 4.022211907095089e-05,
553
+ "loss": 1.631,
554
+ "step": 45500
555
+ },
556
+ {
557
+ "epoch": 0.59,
558
+ "learning_rate": 4.011466982997232e-05,
559
+ "loss": 1.6407,
560
+ "step": 46000
561
+ },
562
+ {
563
+ "epoch": 0.6,
564
+ "learning_rate": 4.000722058899376e-05,
565
+ "loss": 1.6102,
566
+ "step": 46500
567
+ },
568
+ {
569
+ "epoch": 0.61,
570
+ "learning_rate": 3.98997713480152e-05,
571
+ "loss": 1.6339,
572
+ "step": 47000
573
+ },
574
+ {
575
+ "epoch": 0.61,
576
+ "learning_rate": 3.979232210703664e-05,
577
+ "loss": 1.6318,
578
+ "step": 47500
579
+ },
580
+ {
581
+ "epoch": 0.62,
582
+ "learning_rate": 3.968487286605807e-05,
583
+ "loss": 1.6404,
584
+ "step": 48000
585
+ },
586
+ {
587
+ "epoch": 0.63,
588
+ "learning_rate": 3.957742362507951e-05,
589
+ "loss": 1.6324,
590
+ "step": 48500
591
+ },
592
+ {
593
+ "epoch": 0.63,
594
+ "learning_rate": 3.9469974384100956e-05,
595
+ "loss": 1.6366,
596
+ "step": 49000
597
+ },
598
+ {
599
+ "epoch": 0.64,
600
+ "learning_rate": 3.9362525143122394e-05,
601
+ "loss": 1.6355,
602
+ "step": 49500
603
+ },
604
+ {
605
+ "epoch": 0.64,
606
+ "learning_rate": 3.9255075902143826e-05,
607
+ "loss": 1.6333,
608
+ "step": 50000
609
+ },
610
+ {
611
+ "epoch": 0.65,
612
+ "learning_rate": 3.9147626661165264e-05,
613
+ "loss": 1.6354,
614
+ "step": 50500
615
+ },
616
+ {
617
+ "epoch": 0.66,
618
+ "learning_rate": 3.904017742018671e-05,
619
+ "loss": 1.6065,
620
+ "step": 51000
621
+ },
622
+ {
623
+ "epoch": 0.66,
624
+ "learning_rate": 3.893272817920815e-05,
625
+ "loss": 1.6295,
626
+ "step": 51500
627
+ },
628
+ {
629
+ "epoch": 0.67,
630
+ "learning_rate": 3.882527893822958e-05,
631
+ "loss": 1.6177,
632
+ "step": 52000
633
+ },
634
+ {
635
+ "epoch": 0.68,
636
+ "learning_rate": 3.871782969725102e-05,
637
+ "loss": 1.6471,
638
+ "step": 52500
639
+ },
640
+ {
641
+ "epoch": 0.68,
642
+ "learning_rate": 3.861038045627246e-05,
643
+ "loss": 1.6315,
644
+ "step": 53000
645
+ },
646
+ {
647
+ "epoch": 0.69,
648
+ "learning_rate": 3.85029312152939e-05,
649
+ "loss": 1.6103,
650
+ "step": 53500
651
+ },
652
+ {
653
+ "epoch": 0.7,
654
+ "learning_rate": 3.839548197431533e-05,
655
+ "loss": 1.6208,
656
+ "step": 54000
657
+ },
658
+ {
659
+ "epoch": 0.7,
660
+ "learning_rate": 3.8288032733336776e-05,
661
+ "loss": 1.636,
662
+ "step": 54500
663
+ },
664
+ {
665
+ "epoch": 0.71,
666
+ "learning_rate": 3.8180583492358214e-05,
667
+ "loss": 1.6258,
668
+ "step": 55000
669
+ },
670
+ {
671
+ "epoch": 0.72,
672
+ "learning_rate": 3.807313425137965e-05,
673
+ "loss": 1.6249,
674
+ "step": 55500
675
+ },
676
+ {
677
+ "epoch": 0.72,
678
+ "learning_rate": 3.7965685010401084e-05,
679
+ "loss": 1.5984,
680
+ "step": 56000
681
+ },
682
+ {
683
+ "epoch": 0.73,
684
+ "learning_rate": 3.785823576942253e-05,
685
+ "loss": 1.6305,
686
+ "step": 56500
687
+ },
688
+ {
689
+ "epoch": 0.73,
690
+ "learning_rate": 3.775078652844397e-05,
691
+ "loss": 1.6113,
692
+ "step": 57000
693
+ },
694
+ {
695
+ "epoch": 0.74,
696
+ "learning_rate": 3.7643337287465405e-05,
697
+ "loss": 1.6172,
698
+ "step": 57500
699
+ },
700
+ {
701
+ "epoch": 0.75,
702
+ "learning_rate": 3.7535888046486837e-05,
703
+ "loss": 1.6103,
704
+ "step": 58000
705
+ },
706
+ {
707
+ "epoch": 0.75,
708
+ "learning_rate": 3.742843880550828e-05,
709
+ "loss": 1.6217,
710
+ "step": 58500
711
+ },
712
+ {
713
+ "epoch": 0.76,
714
+ "learning_rate": 3.732098956452972e-05,
715
+ "loss": 1.626,
716
+ "step": 59000
717
+ },
718
+ {
719
+ "epoch": 0.77,
720
+ "learning_rate": 3.721354032355116e-05,
721
+ "loss": 1.6133,
722
+ "step": 59500
723
+ },
724
+ {
725
+ "epoch": 0.77,
726
+ "learning_rate": 3.7106091082572596e-05,
727
+ "loss": 1.592,
728
+ "step": 60000
729
+ },
730
+ {
731
+ "epoch": 0.78,
732
+ "learning_rate": 3.6998641841594034e-05,
733
+ "loss": 1.6061,
734
+ "step": 60500
735
+ },
736
+ {
737
+ "epoch": 0.79,
738
+ "learning_rate": 3.689119260061547e-05,
739
+ "loss": 1.5942,
740
+ "step": 61000
741
+ },
742
+ {
743
+ "epoch": 0.79,
744
+ "learning_rate": 3.678374335963691e-05,
745
+ "loss": 1.6096,
746
+ "step": 61500
747
+ },
748
+ {
749
+ "epoch": 0.8,
750
+ "learning_rate": 3.667629411865835e-05,
751
+ "loss": 1.6069,
752
+ "step": 62000
753
+ },
754
+ {
755
+ "epoch": 0.81,
756
+ "learning_rate": 3.656884487767979e-05,
757
+ "loss": 1.6086,
758
+ "step": 62500
759
+ },
760
+ {
761
+ "epoch": 0.81,
762
+ "learning_rate": 3.6461395636701225e-05,
763
+ "loss": 1.6015,
764
+ "step": 63000
765
+ },
766
+ {
767
+ "epoch": 0.82,
768
+ "learning_rate": 3.635394639572266e-05,
769
+ "loss": 1.602,
770
+ "step": 63500
771
+ },
772
+ {
773
+ "epoch": 0.83,
774
+ "learning_rate": 3.62464971547441e-05,
775
+ "loss": 1.6043,
776
+ "step": 64000
777
+ },
778
+ {
779
+ "epoch": 0.83,
780
+ "learning_rate": 3.613904791376554e-05,
781
+ "loss": 1.6037,
782
+ "step": 64500
783
+ },
784
+ {
785
+ "epoch": 0.84,
786
+ "learning_rate": 3.603159867278698e-05,
787
+ "loss": 1.6092,
788
+ "step": 65000
789
+ },
790
+ {
791
+ "epoch": 0.84,
792
+ "learning_rate": 3.5924149431808416e-05,
793
+ "loss": 1.6197,
794
+ "step": 65500
795
+ },
796
+ {
797
+ "epoch": 0.85,
798
+ "learning_rate": 3.5816700190829854e-05,
799
+ "loss": 1.5959,
800
+ "step": 66000
801
+ },
802
+ {
803
+ "epoch": 0.86,
804
+ "learning_rate": 3.570925094985129e-05,
805
+ "loss": 1.5951,
806
+ "step": 66500
807
+ },
808
+ {
809
+ "epoch": 0.86,
810
+ "learning_rate": 3.560180170887273e-05,
811
+ "loss": 1.5979,
812
+ "step": 67000
813
+ },
814
+ {
815
+ "epoch": 0.87,
816
+ "learning_rate": 3.549435246789417e-05,
817
+ "loss": 1.6053,
818
+ "step": 67500
819
+ },
820
+ {
821
+ "epoch": 0.88,
822
+ "learning_rate": 3.538690322691561e-05,
823
+ "loss": 1.5998,
824
+ "step": 68000
825
+ },
826
+ {
827
+ "epoch": 0.88,
828
+ "learning_rate": 3.5279453985937045e-05,
829
+ "loss": 1.6112,
830
+ "step": 68500
831
+ },
832
+ {
833
+ "epoch": 0.89,
834
+ "learning_rate": 3.517200474495848e-05,
835
+ "loss": 1.5938,
836
+ "step": 69000
837
+ },
838
+ {
839
+ "epoch": 0.9,
840
+ "learning_rate": 3.506455550397992e-05,
841
+ "loss": 1.6034,
842
+ "step": 69500
843
+ },
844
+ {
845
+ "epoch": 0.9,
846
+ "learning_rate": 3.495710626300136e-05,
847
+ "loss": 1.6093,
848
+ "step": 70000
849
+ },
850
+ {
851
+ "epoch": 0.91,
852
+ "learning_rate": 3.48496570220228e-05,
853
+ "loss": 1.6027,
854
+ "step": 70500
855
+ },
856
+ {
857
+ "epoch": 0.92,
858
+ "learning_rate": 3.4742207781044236e-05,
859
+ "loss": 1.6009,
860
+ "step": 71000
861
+ },
862
+ {
863
+ "epoch": 0.92,
864
+ "learning_rate": 3.4634758540065674e-05,
865
+ "loss": 1.5985,
866
+ "step": 71500
867
+ },
868
+ {
869
+ "epoch": 0.93,
870
+ "learning_rate": 3.452730929908711e-05,
871
+ "loss": 1.607,
872
+ "step": 72000
873
+ },
874
+ {
875
+ "epoch": 0.93,
876
+ "learning_rate": 3.441986005810855e-05,
877
+ "loss": 1.5853,
878
+ "step": 72500
879
+ },
880
+ {
881
+ "epoch": 0.94,
882
+ "learning_rate": 3.4312410817129995e-05,
883
+ "loss": 1.5946,
884
+ "step": 73000
885
+ },
886
+ {
887
+ "epoch": 0.95,
888
+ "learning_rate": 3.4204961576151426e-05,
889
+ "loss": 1.5972,
890
+ "step": 73500
891
+ },
892
+ {
893
+ "epoch": 0.95,
894
+ "learning_rate": 3.4097512335172865e-05,
895
+ "loss": 1.5912,
896
+ "step": 74000
897
+ },
898
+ {
899
+ "epoch": 0.96,
900
+ "learning_rate": 3.39900630941943e-05,
901
+ "loss": 1.591,
902
+ "step": 74500
903
+ },
904
+ {
905
+ "epoch": 0.97,
906
+ "learning_rate": 3.388261385321575e-05,
907
+ "loss": 1.5981,
908
+ "step": 75000
909
+ },
910
+ {
911
+ "epoch": 0.97,
912
+ "learning_rate": 3.377516461223718e-05,
913
+ "loss": 1.5897,
914
+ "step": 75500
915
+ },
916
+ {
917
+ "epoch": 0.98,
918
+ "learning_rate": 3.366771537125862e-05,
919
+ "loss": 1.588,
920
+ "step": 76000
921
+ },
922
+ {
923
+ "epoch": 0.99,
924
+ "learning_rate": 3.3560266130280055e-05,
925
+ "loss": 1.5872,
926
+ "step": 76500
927
+ },
928
+ {
929
+ "epoch": 0.99,
930
+ "learning_rate": 3.34528168893015e-05,
931
+ "loss": 1.593,
932
+ "step": 77000
933
+ },
934
+ {
935
+ "epoch": 1.0,
936
+ "learning_rate": 3.334536764832293e-05,
937
+ "loss": 1.588,
938
+ "step": 77500
939
+ },
940
+ {
941
+ "epoch": 1.01,
942
+ "learning_rate": 3.323791840734437e-05,
943
+ "loss": 1.5065,
944
+ "step": 78000
945
+ },
946
+ {
947
+ "epoch": 1.01,
948
+ "learning_rate": 3.3130469166365815e-05,
949
+ "loss": 1.486,
950
+ "step": 78500
951
+ },
952
+ {
953
+ "epoch": 1.02,
954
+ "learning_rate": 3.3023019925387246e-05,
955
+ "loss": 1.4949,
956
+ "step": 79000
957
+ },
958
+ {
959
+ "epoch": 1.03,
960
+ "learning_rate": 3.2915570684408684e-05,
961
+ "loss": 1.4824,
962
+ "step": 79500
963
+ },
964
+ {
965
+ "epoch": 1.03,
966
+ "learning_rate": 3.280812144343012e-05,
967
+ "loss": 1.4943,
968
+ "step": 80000
969
+ },
970
+ {
971
+ "epoch": 1.04,
972
+ "learning_rate": 3.270067220245157e-05,
973
+ "loss": 1.489,
974
+ "step": 80500
975
+ },
976
+ {
977
+ "epoch": 1.04,
978
+ "learning_rate": 3.2593222961473e-05,
979
+ "loss": 1.4835,
980
+ "step": 81000
981
+ },
982
+ {
983
+ "epoch": 1.05,
984
+ "learning_rate": 3.248577372049444e-05,
985
+ "loss": 1.4919,
986
+ "step": 81500
987
+ },
988
+ {
989
+ "epoch": 1.06,
990
+ "learning_rate": 3.2378324479515875e-05,
991
+ "loss": 1.4934,
992
+ "step": 82000
993
+ },
994
+ {
995
+ "epoch": 1.06,
996
+ "learning_rate": 3.227087523853732e-05,
997
+ "loss": 1.4846,
998
+ "step": 82500
999
+ },
1000
+ {
1001
+ "epoch": 1.07,
1002
+ "learning_rate": 3.216342599755875e-05,
1003
+ "loss": 1.4969,
1004
+ "step": 83000
1005
+ },
1006
+ {
1007
+ "epoch": 1.08,
1008
+ "learning_rate": 3.205597675658019e-05,
1009
+ "loss": 1.4921,
1010
+ "step": 83500
1011
+ },
1012
+ {
1013
+ "epoch": 1.08,
1014
+ "learning_rate": 3.1948527515601635e-05,
1015
+ "loss": 1.4874,
1016
+ "step": 84000
1017
+ },
1018
+ {
1019
+ "epoch": 1.09,
1020
+ "learning_rate": 3.184107827462307e-05,
1021
+ "loss": 1.513,
1022
+ "step": 84500
1023
+ },
1024
+ {
1025
+ "epoch": 1.1,
1026
+ "learning_rate": 3.1733629033644504e-05,
1027
+ "loss": 1.5016,
1028
+ "step": 85000
1029
+ },
1030
+ {
1031
+ "epoch": 1.1,
1032
+ "learning_rate": 3.162617979266594e-05,
1033
+ "loss": 1.4884,
1034
+ "step": 85500
1035
+ },
1036
+ {
1037
+ "epoch": 1.11,
1038
+ "learning_rate": 3.151873055168739e-05,
1039
+ "loss": 1.5143,
1040
+ "step": 86000
1041
+ },
1042
+ {
1043
+ "epoch": 1.12,
1044
+ "learning_rate": 3.1411281310708826e-05,
1045
+ "loss": 1.4998,
1046
+ "step": 86500
1047
+ },
1048
+ {
1049
+ "epoch": 1.12,
1050
+ "learning_rate": 3.130383206973026e-05,
1051
+ "loss": 1.4906,
1052
+ "step": 87000
1053
+ },
1054
+ {
1055
+ "epoch": 1.13,
1056
+ "learning_rate": 3.1196382828751695e-05,
1057
+ "loss": 1.4958,
1058
+ "step": 87500
1059
+ },
1060
+ {
1061
+ "epoch": 1.13,
1062
+ "learning_rate": 3.108893358777314e-05,
1063
+ "loss": 1.4954,
1064
+ "step": 88000
1065
+ },
1066
+ {
1067
+ "epoch": 1.14,
1068
+ "learning_rate": 3.098148434679458e-05,
1069
+ "loss": 1.4951,
1070
+ "step": 88500
1071
+ },
1072
+ {
1073
+ "epoch": 1.15,
1074
+ "learning_rate": 3.087403510581601e-05,
1075
+ "loss": 1.4895,
1076
+ "step": 89000
1077
+ },
1078
+ {
1079
+ "epoch": 1.15,
1080
+ "learning_rate": 3.0766585864837455e-05,
1081
+ "loss": 1.4779,
1082
+ "step": 89500
1083
+ },
1084
+ {
1085
+ "epoch": 1.16,
1086
+ "learning_rate": 3.065913662385889e-05,
1087
+ "loss": 1.4993,
1088
+ "step": 90000
1089
+ },
1090
+ {
1091
+ "epoch": 1.17,
1092
+ "learning_rate": 3.055168738288033e-05,
1093
+ "loss": 1.4942,
1094
+ "step": 90500
1095
+ },
1096
+ {
1097
+ "epoch": 1.17,
1098
+ "learning_rate": 3.0444238141901766e-05,
1099
+ "loss": 1.4917,
1100
+ "step": 91000
1101
+ },
1102
+ {
1103
+ "epoch": 1.18,
1104
+ "learning_rate": 3.0336788900923207e-05,
1105
+ "loss": 1.5005,
1106
+ "step": 91500
1107
+ },
1108
+ {
1109
+ "epoch": 1.19,
1110
+ "learning_rate": 3.0229339659944645e-05,
1111
+ "loss": 1.4911,
1112
+ "step": 92000
1113
+ },
1114
+ {
1115
+ "epoch": 1.19,
1116
+ "learning_rate": 3.012189041896608e-05,
1117
+ "loss": 1.4791,
1118
+ "step": 92500
1119
+ },
1120
+ {
1121
+ "epoch": 1.2,
1122
+ "learning_rate": 3.001444117798752e-05,
1123
+ "loss": 1.4972,
1124
+ "step": 93000
1125
+ },
1126
+ {
1127
+ "epoch": 1.21,
1128
+ "learning_rate": 2.990699193700896e-05,
1129
+ "loss": 1.5062,
1130
+ "step": 93500
1131
+ },
1132
+ {
1133
+ "epoch": 1.21,
1134
+ "learning_rate": 2.9799542696030398e-05,
1135
+ "loss": 1.4845,
1136
+ "step": 94000
1137
+ },
1138
+ {
1139
+ "epoch": 1.22,
1140
+ "learning_rate": 2.9692093455051833e-05,
1141
+ "loss": 1.507,
1142
+ "step": 94500
1143
+ },
1144
+ {
1145
+ "epoch": 1.22,
1146
+ "learning_rate": 2.958464421407327e-05,
1147
+ "loss": 1.4963,
1148
+ "step": 95000
1149
+ },
1150
+ {
1151
+ "epoch": 1.23,
1152
+ "learning_rate": 2.9477194973094713e-05,
1153
+ "loss": 1.4845,
1154
+ "step": 95500
1155
+ },
1156
+ {
1157
+ "epoch": 1.24,
1158
+ "learning_rate": 2.936974573211615e-05,
1159
+ "loss": 1.5025,
1160
+ "step": 96000
1161
+ },
1162
+ {
1163
+ "epoch": 1.24,
1164
+ "learning_rate": 2.9262296491137586e-05,
1165
+ "loss": 1.4935,
1166
+ "step": 96500
1167
+ },
1168
+ {
1169
+ "epoch": 1.25,
1170
+ "learning_rate": 2.9154847250159027e-05,
1171
+ "loss": 1.4881,
1172
+ "step": 97000
1173
+ },
1174
+ {
1175
+ "epoch": 1.26,
1176
+ "learning_rate": 2.9047398009180465e-05,
1177
+ "loss": 1.5061,
1178
+ "step": 97500
1179
+ },
1180
+ {
1181
+ "epoch": 1.26,
1182
+ "learning_rate": 2.8939948768201903e-05,
1183
+ "loss": 1.5015,
1184
+ "step": 98000
1185
+ },
1186
+ {
1187
+ "epoch": 1.27,
1188
+ "learning_rate": 2.8832499527223338e-05,
1189
+ "loss": 1.4909,
1190
+ "step": 98500
1191
+ },
1192
+ {
1193
+ "epoch": 1.28,
1194
+ "learning_rate": 2.872505028624478e-05,
1195
+ "loss": 1.496,
1196
+ "step": 99000
1197
+ },
1198
+ {
1199
+ "epoch": 1.28,
1200
+ "learning_rate": 2.8617601045266218e-05,
1201
+ "loss": 1.4906,
1202
+ "step": 99500
1203
+ },
1204
+ {
1205
+ "epoch": 1.29,
1206
+ "learning_rate": 2.8510151804287656e-05,
1207
+ "loss": 1.4941,
1208
+ "step": 100000
1209
+ },
1210
+ {
1211
+ "epoch": 1.3,
1212
+ "learning_rate": 2.840270256330909e-05,
1213
+ "loss": 1.5103,
1214
+ "step": 100500
1215
+ },
1216
+ {
1217
+ "epoch": 1.3,
1218
+ "learning_rate": 2.8295253322330532e-05,
1219
+ "loss": 1.4857,
1220
+ "step": 101000
1221
+ },
1222
+ {
1223
+ "epoch": 1.31,
1224
+ "learning_rate": 2.818780408135197e-05,
1225
+ "loss": 1.503,
1226
+ "step": 101500
1227
+ },
1228
+ {
1229
+ "epoch": 1.32,
1230
+ "learning_rate": 2.808035484037341e-05,
1231
+ "loss": 1.5008,
1232
+ "step": 102000
1233
+ },
1234
+ {
1235
+ "epoch": 1.32,
1236
+ "learning_rate": 2.797290559939485e-05,
1237
+ "loss": 1.4872,
1238
+ "step": 102500
1239
+ },
1240
+ {
1241
+ "epoch": 1.33,
1242
+ "learning_rate": 2.7865456358416285e-05,
1243
+ "loss": 1.4949,
1244
+ "step": 103000
1245
+ },
1246
+ {
1247
+ "epoch": 1.33,
1248
+ "learning_rate": 2.7758007117437723e-05,
1249
+ "loss": 1.4947,
1250
+ "step": 103500
1251
+ },
1252
+ {
1253
+ "epoch": 1.34,
1254
+ "learning_rate": 2.765055787645916e-05,
1255
+ "loss": 1.4965,
1256
+ "step": 104000
1257
+ },
1258
+ {
1259
+ "epoch": 1.35,
1260
+ "learning_rate": 2.7543108635480603e-05,
1261
+ "loss": 1.5043,
1262
+ "step": 104500
1263
+ },
1264
+ {
1265
+ "epoch": 1.35,
1266
+ "learning_rate": 2.7435659394502038e-05,
1267
+ "loss": 1.5026,
1268
+ "step": 105000
1269
+ },
1270
+ {
1271
+ "epoch": 1.36,
1272
+ "learning_rate": 2.7328210153523476e-05,
1273
+ "loss": 1.4921,
1274
+ "step": 105500
1275
+ },
1276
+ {
1277
+ "epoch": 1.37,
1278
+ "learning_rate": 2.7220760912544914e-05,
1279
+ "loss": 1.4981,
1280
+ "step": 106000
1281
+ },
1282
+ {
1283
+ "epoch": 1.37,
1284
+ "learning_rate": 2.7113311671566356e-05,
1285
+ "loss": 1.4852,
1286
+ "step": 106500
1287
+ },
1288
+ {
1289
+ "epoch": 1.38,
1290
+ "learning_rate": 2.700586243058779e-05,
1291
+ "loss": 1.4993,
1292
+ "step": 107000
1293
+ },
1294
+ {
1295
+ "epoch": 1.39,
1296
+ "learning_rate": 2.689841318960923e-05,
1297
+ "loss": 1.4835,
1298
+ "step": 107500
1299
+ },
1300
+ {
1301
+ "epoch": 1.39,
1302
+ "learning_rate": 2.679096394863067e-05,
1303
+ "loss": 1.499,
1304
+ "step": 108000
1305
+ },
1306
+ {
1307
+ "epoch": 1.4,
1308
+ "learning_rate": 2.6683514707652108e-05,
1309
+ "loss": 1.4875,
1310
+ "step": 108500
1311
+ },
1312
+ {
1313
+ "epoch": 1.41,
1314
+ "learning_rate": 2.6576065466673543e-05,
1315
+ "loss": 1.4828,
1316
+ "step": 109000
1317
+ },
1318
+ {
1319
+ "epoch": 1.41,
1320
+ "learning_rate": 2.646861622569498e-05,
1321
+ "loss": 1.4887,
1322
+ "step": 109500
1323
+ },
1324
+ {
1325
+ "epoch": 1.42,
1326
+ "learning_rate": 2.6361166984716423e-05,
1327
+ "loss": 1.4869,
1328
+ "step": 110000
1329
+ },
1330
+ {
1331
+ "epoch": 1.42,
1332
+ "learning_rate": 2.625371774373786e-05,
1333
+ "loss": 1.4809,
1334
+ "step": 110500
1335
+ },
1336
+ {
1337
+ "epoch": 1.43,
1338
+ "learning_rate": 2.6146268502759296e-05,
1339
+ "loss": 1.4856,
1340
+ "step": 111000
1341
+ },
1342
+ {
1343
+ "epoch": 1.44,
1344
+ "learning_rate": 2.6038819261780734e-05,
1345
+ "loss": 1.4923,
1346
+ "step": 111500
1347
+ },
1348
+ {
1349
+ "epoch": 1.44,
1350
+ "learning_rate": 2.5931370020802175e-05,
1351
+ "loss": 1.4985,
1352
+ "step": 112000
1353
+ },
1354
+ {
1355
+ "epoch": 1.45,
1356
+ "learning_rate": 2.5823920779823614e-05,
1357
+ "loss": 1.4979,
1358
+ "step": 112500
1359
+ },
1360
+ {
1361
+ "epoch": 1.46,
1362
+ "learning_rate": 2.571647153884505e-05,
1363
+ "loss": 1.5053,
1364
+ "step": 113000
1365
+ },
1366
+ {
1367
+ "epoch": 1.46,
1368
+ "learning_rate": 2.5609022297866493e-05,
1369
+ "loss": 1.4821,
1370
+ "step": 113500
1371
+ },
1372
+ {
1373
+ "epoch": 1.47,
1374
+ "learning_rate": 2.5501573056887928e-05,
1375
+ "loss": 1.4847,
1376
+ "step": 114000
1377
+ },
1378
+ {
1379
+ "epoch": 1.48,
1380
+ "learning_rate": 2.5394123815909366e-05,
1381
+ "loss": 1.4815,
1382
+ "step": 114500
1383
+ },
1384
+ {
1385
+ "epoch": 1.48,
1386
+ "learning_rate": 2.52866745749308e-05,
1387
+ "loss": 1.4723,
1388
+ "step": 115000
1389
+ },
1390
+ {
1391
+ "epoch": 1.49,
1392
+ "learning_rate": 2.5179225333952246e-05,
1393
+ "loss": 1.4806,
1394
+ "step": 115500
1395
+ },
1396
+ {
1397
+ "epoch": 1.5,
1398
+ "learning_rate": 2.507177609297368e-05,
1399
+ "loss": 1.4784,
1400
+ "step": 116000
1401
+ },
1402
+ {
1403
+ "epoch": 1.5,
1404
+ "learning_rate": 2.496432685199512e-05,
1405
+ "loss": 1.4809,
1406
+ "step": 116500
1407
+ },
1408
+ {
1409
+ "epoch": 1.51,
1410
+ "learning_rate": 2.4856877611016557e-05,
1411
+ "loss": 1.4832,
1412
+ "step": 117000
1413
+ },
1414
+ {
1415
+ "epoch": 1.52,
1416
+ "learning_rate": 2.4749428370037995e-05,
1417
+ "loss": 1.471,
1418
+ "step": 117500
1419
+ },
1420
+ {
1421
+ "epoch": 1.52,
1422
+ "learning_rate": 2.4641979129059433e-05,
1423
+ "loss": 1.5024,
1424
+ "step": 118000
1425
+ },
1426
+ {
1427
+ "epoch": 1.53,
1428
+ "learning_rate": 2.453452988808087e-05,
1429
+ "loss": 1.489,
1430
+ "step": 118500
1431
+ },
1432
+ {
1433
+ "epoch": 1.53,
1434
+ "learning_rate": 2.442708064710231e-05,
1435
+ "loss": 1.4916,
1436
+ "step": 119000
1437
+ },
1438
+ {
1439
+ "epoch": 1.54,
1440
+ "learning_rate": 2.4319631406123748e-05,
1441
+ "loss": 1.48,
1442
+ "step": 119500
1443
+ },
1444
+ {
1445
+ "epoch": 1.55,
1446
+ "learning_rate": 2.4212182165145186e-05,
1447
+ "loss": 1.4802,
1448
+ "step": 120000
1449
+ },
1450
+ {
1451
+ "epoch": 1.55,
1452
+ "learning_rate": 2.4104732924166624e-05,
1453
+ "loss": 1.4931,
1454
+ "step": 120500
1455
+ },
1456
+ {
1457
+ "epoch": 1.56,
1458
+ "learning_rate": 2.3997283683188062e-05,
1459
+ "loss": 1.4738,
1460
+ "step": 121000
1461
+ },
1462
+ {
1463
+ "epoch": 1.57,
1464
+ "learning_rate": 2.38898344422095e-05,
1465
+ "loss": 1.4885,
1466
+ "step": 121500
1467
+ },
1468
+ {
1469
+ "epoch": 1.57,
1470
+ "learning_rate": 2.378238520123094e-05,
1471
+ "loss": 1.4863,
1472
+ "step": 122000
1473
+ },
1474
+ {
1475
+ "epoch": 1.58,
1476
+ "learning_rate": 2.3674935960252377e-05,
1477
+ "loss": 1.4862,
1478
+ "step": 122500
1479
+ },
1480
+ {
1481
+ "epoch": 1.59,
1482
+ "learning_rate": 2.3567486719273815e-05,
1483
+ "loss": 1.4913,
1484
+ "step": 123000
1485
+ },
1486
+ {
1487
+ "epoch": 1.59,
1488
+ "learning_rate": 2.3460037478295253e-05,
1489
+ "loss": 1.4889,
1490
+ "step": 123500
1491
+ },
1492
+ {
1493
+ "epoch": 1.6,
1494
+ "learning_rate": 2.335258823731669e-05,
1495
+ "loss": 1.4756,
1496
+ "step": 124000
1497
+ },
1498
+ {
1499
+ "epoch": 1.61,
1500
+ "learning_rate": 2.324513899633813e-05,
1501
+ "loss": 1.4789,
1502
+ "step": 124500
1503
+ },
1504
+ {
1505
+ "epoch": 1.61,
1506
+ "learning_rate": 2.313768975535957e-05,
1507
+ "loss": 1.4708,
1508
+ "step": 125000
1509
+ },
1510
+ {
1511
+ "epoch": 1.62,
1512
+ "learning_rate": 2.3030240514381006e-05,
1513
+ "loss": 1.4923,
1514
+ "step": 125500
1515
+ },
1516
+ {
1517
+ "epoch": 1.62,
1518
+ "learning_rate": 2.2922791273402448e-05,
1519
+ "loss": 1.4843,
1520
+ "step": 126000
1521
+ },
1522
+ {
1523
+ "epoch": 1.63,
1524
+ "learning_rate": 2.2815342032423882e-05,
1525
+ "loss": 1.4844,
1526
+ "step": 126500
1527
+ },
1528
+ {
1529
+ "epoch": 1.64,
1530
+ "learning_rate": 2.2707892791445324e-05,
1531
+ "loss": 1.4778,
1532
+ "step": 127000
1533
+ },
1534
+ {
1535
+ "epoch": 1.64,
1536
+ "learning_rate": 2.260044355046676e-05,
1537
+ "loss": 1.4865,
1538
+ "step": 127500
1539
+ },
1540
+ {
1541
+ "epoch": 1.65,
1542
+ "learning_rate": 2.24929943094882e-05,
1543
+ "loss": 1.4807,
1544
+ "step": 128000
1545
+ },
1546
+ {
1547
+ "epoch": 1.66,
1548
+ "learning_rate": 2.2385545068509635e-05,
1549
+ "loss": 1.4846,
1550
+ "step": 128500
1551
+ },
1552
+ {
1553
+ "epoch": 1.66,
1554
+ "learning_rate": 2.2278095827531077e-05,
1555
+ "loss": 1.4882,
1556
+ "step": 129000
1557
+ },
1558
+ {
1559
+ "epoch": 1.67,
1560
+ "learning_rate": 2.217064658655251e-05,
1561
+ "loss": 1.4725,
1562
+ "step": 129500
1563
+ },
1564
+ {
1565
+ "epoch": 1.68,
1566
+ "learning_rate": 2.2063197345573953e-05,
1567
+ "loss": 1.4963,
1568
+ "step": 130000
1569
+ },
1570
+ {
1571
+ "epoch": 1.68,
1572
+ "learning_rate": 2.1955748104595388e-05,
1573
+ "loss": 1.4714,
1574
+ "step": 130500
1575
+ },
1576
+ {
1577
+ "epoch": 1.69,
1578
+ "learning_rate": 2.184829886361683e-05,
1579
+ "loss": 1.4767,
1580
+ "step": 131000
1581
+ },
1582
+ {
1583
+ "epoch": 1.7,
1584
+ "learning_rate": 2.1740849622638267e-05,
1585
+ "loss": 1.4821,
1586
+ "step": 131500
1587
+ },
1588
+ {
1589
+ "epoch": 1.7,
1590
+ "learning_rate": 2.1633400381659706e-05,
1591
+ "loss": 1.4827,
1592
+ "step": 132000
1593
+ },
1594
+ {
1595
+ "epoch": 1.71,
1596
+ "learning_rate": 2.1525951140681144e-05,
1597
+ "loss": 1.4768,
1598
+ "step": 132500
1599
+ },
1600
+ {
1601
+ "epoch": 1.71,
1602
+ "learning_rate": 2.1418501899702582e-05,
1603
+ "loss": 1.4739,
1604
+ "step": 133000
1605
+ },
1606
+ {
1607
+ "epoch": 1.72,
1608
+ "learning_rate": 2.131105265872402e-05,
1609
+ "loss": 1.4644,
1610
+ "step": 133500
1611
+ },
1612
+ {
1613
+ "epoch": 1.73,
1614
+ "learning_rate": 2.1203603417745458e-05,
1615
+ "loss": 1.4779,
1616
+ "step": 134000
1617
+ },
1618
+ {
1619
+ "epoch": 1.73,
1620
+ "learning_rate": 2.1096154176766896e-05,
1621
+ "loss": 1.479,
1622
+ "step": 134500
1623
+ },
1624
+ {
1625
+ "epoch": 1.74,
1626
+ "learning_rate": 2.0988704935788335e-05,
1627
+ "loss": 1.4668,
1628
+ "step": 135000
1629
+ },
1630
+ {
1631
+ "epoch": 1.75,
1632
+ "learning_rate": 2.0881255694809773e-05,
1633
+ "loss": 1.4794,
1634
+ "step": 135500
1635
+ },
1636
+ {
1637
+ "epoch": 1.75,
1638
+ "learning_rate": 2.077380645383121e-05,
1639
+ "loss": 1.4734,
1640
+ "step": 136000
1641
+ },
1642
+ {
1643
+ "epoch": 1.76,
1644
+ "learning_rate": 2.066635721285265e-05,
1645
+ "loss": 1.4826,
1646
+ "step": 136500
1647
+ },
1648
+ {
1649
+ "epoch": 1.77,
1650
+ "learning_rate": 2.055890797187409e-05,
1651
+ "loss": 1.4691,
1652
+ "step": 137000
1653
+ },
1654
+ {
1655
+ "epoch": 1.77,
1656
+ "learning_rate": 2.0451458730895525e-05,
1657
+ "loss": 1.4846,
1658
+ "step": 137500
1659
+ },
1660
+ {
1661
+ "epoch": 1.78,
1662
+ "learning_rate": 2.0344009489916967e-05,
1663
+ "loss": 1.4644,
1664
+ "step": 138000
1665
+ },
1666
+ {
1667
+ "epoch": 1.79,
1668
+ "learning_rate": 2.02365602489384e-05,
1669
+ "loss": 1.4655,
1670
+ "step": 138500
1671
+ },
1672
+ {
1673
+ "epoch": 1.79,
1674
+ "learning_rate": 2.0129111007959843e-05,
1675
+ "loss": 1.4749,
1676
+ "step": 139000
1677
+ },
1678
+ {
1679
+ "epoch": 1.8,
1680
+ "learning_rate": 2.0021661766981278e-05,
1681
+ "loss": 1.4665,
1682
+ "step": 139500
1683
+ },
1684
+ {
1685
+ "epoch": 1.81,
1686
+ "learning_rate": 1.991421252600272e-05,
1687
+ "loss": 1.4736,
1688
+ "step": 140000
1689
+ },
1690
+ {
1691
+ "epoch": 1.81,
1692
+ "learning_rate": 1.9806763285024154e-05,
1693
+ "loss": 1.4729,
1694
+ "step": 140500
1695
+ },
1696
+ {
1697
+ "epoch": 1.82,
1698
+ "learning_rate": 1.9699314044045593e-05,
1699
+ "loss": 1.4615,
1700
+ "step": 141000
1701
+ },
1702
+ {
1703
+ "epoch": 1.82,
1704
+ "learning_rate": 1.959186480306703e-05,
1705
+ "loss": 1.4788,
1706
+ "step": 141500
1707
+ },
1708
+ {
1709
+ "epoch": 1.83,
1710
+ "learning_rate": 1.948441556208847e-05,
1711
+ "loss": 1.4687,
1712
+ "step": 142000
1713
+ },
1714
+ {
1715
+ "epoch": 1.84,
1716
+ "learning_rate": 1.9376966321109907e-05,
1717
+ "loss": 1.4595,
1718
+ "step": 142500
1719
+ },
1720
+ {
1721
+ "epoch": 1.84,
1722
+ "learning_rate": 1.9269517080131345e-05,
1723
+ "loss": 1.4754,
1724
+ "step": 143000
1725
+ },
1726
+ {
1727
+ "epoch": 1.85,
1728
+ "learning_rate": 1.9162067839152787e-05,
1729
+ "loss": 1.4687,
1730
+ "step": 143500
1731
+ },
1732
+ {
1733
+ "epoch": 1.86,
1734
+ "learning_rate": 1.905461859817422e-05,
1735
+ "loss": 1.4911,
1736
+ "step": 144000
1737
+ },
1738
+ {
1739
+ "epoch": 1.86,
1740
+ "learning_rate": 1.8947169357195663e-05,
1741
+ "loss": 1.4912,
1742
+ "step": 144500
1743
+ },
1744
+ {
1745
+ "epoch": 1.87,
1746
+ "learning_rate": 1.8839720116217098e-05,
1747
+ "loss": 1.4717,
1748
+ "step": 145000
1749
+ },
1750
+ {
1751
+ "epoch": 1.88,
1752
+ "learning_rate": 1.873227087523854e-05,
1753
+ "loss": 1.4774,
1754
+ "step": 145500
1755
+ },
1756
+ {
1757
+ "epoch": 1.88,
1758
+ "learning_rate": 1.8624821634259974e-05,
1759
+ "loss": 1.4643,
1760
+ "step": 146000
1761
+ },
1762
+ {
1763
+ "epoch": 1.89,
1764
+ "learning_rate": 1.8517372393281416e-05,
1765
+ "loss": 1.4708,
1766
+ "step": 146500
1767
+ },
1768
+ {
1769
+ "epoch": 1.9,
1770
+ "learning_rate": 1.840992315230285e-05,
1771
+ "loss": 1.4627,
1772
+ "step": 147000
1773
+ },
1774
+ {
1775
+ "epoch": 1.9,
1776
+ "learning_rate": 1.8302473911324292e-05,
1777
+ "loss": 1.4712,
1778
+ "step": 147500
1779
+ },
1780
+ {
1781
+ "epoch": 1.91,
1782
+ "learning_rate": 1.8195024670345727e-05,
1783
+ "loss": 1.4798,
1784
+ "step": 148000
1785
+ },
1786
+ {
1787
+ "epoch": 1.91,
1788
+ "learning_rate": 1.808757542936717e-05,
1789
+ "loss": 1.4701,
1790
+ "step": 148500
1791
+ },
1792
+ {
1793
+ "epoch": 1.92,
1794
+ "learning_rate": 1.7980126188388607e-05,
1795
+ "loss": 1.4658,
1796
+ "step": 149000
1797
+ },
1798
+ {
1799
+ "epoch": 1.93,
1800
+ "learning_rate": 1.7872676947410045e-05,
1801
+ "loss": 1.4712,
1802
+ "step": 149500
1803
+ },
1804
+ {
1805
+ "epoch": 1.93,
1806
+ "learning_rate": 1.7765227706431483e-05,
1807
+ "loss": 1.4665,
1808
+ "step": 150000
1809
+ },
1810
+ {
1811
+ "epoch": 1.94,
1812
+ "learning_rate": 1.765777846545292e-05,
1813
+ "loss": 1.487,
1814
+ "step": 150500
1815
+ },
1816
+ {
1817
+ "epoch": 1.95,
1818
+ "learning_rate": 1.755032922447436e-05,
1819
+ "loss": 1.4585,
1820
+ "step": 151000
1821
+ },
1822
+ {
1823
+ "epoch": 1.95,
1824
+ "learning_rate": 1.7442879983495797e-05,
1825
+ "loss": 1.4496,
1826
+ "step": 151500
1827
+ },
1828
+ {
1829
+ "epoch": 1.96,
1830
+ "learning_rate": 1.7335430742517236e-05,
1831
+ "loss": 1.4655,
1832
+ "step": 152000
1833
+ },
1834
+ {
1835
+ "epoch": 1.97,
1836
+ "learning_rate": 1.7227981501538674e-05,
1837
+ "loss": 1.4583,
1838
+ "step": 152500
1839
+ },
1840
+ {
1841
+ "epoch": 1.97,
1842
+ "learning_rate": 1.7120532260560112e-05,
1843
+ "loss": 1.4734,
1844
+ "step": 153000
1845
+ },
1846
+ {
1847
+ "epoch": 1.98,
1848
+ "learning_rate": 1.701308301958155e-05,
1849
+ "loss": 1.463,
1850
+ "step": 153500
1851
+ },
1852
+ {
1853
+ "epoch": 1.99,
1854
+ "learning_rate": 1.6905633778602988e-05,
1855
+ "loss": 1.474,
1856
+ "step": 154000
1857
+ },
1858
+ {
1859
+ "epoch": 1.99,
1860
+ "learning_rate": 1.6798184537624426e-05,
1861
+ "loss": 1.4687,
1862
+ "step": 154500
1863
+ },
1864
+ {
1865
+ "epoch": 2.0,
1866
+ "learning_rate": 1.6690735296645865e-05,
1867
+ "loss": 1.4543,
1868
+ "step": 155000
1869
+ },
1870
+ {
1871
+ "epoch": 2.01,
1872
+ "learning_rate": 1.6583286055667306e-05,
1873
+ "loss": 1.3815,
1874
+ "step": 155500
1875
+ },
1876
+ {
1877
+ "epoch": 2.01,
1878
+ "learning_rate": 1.647583681468874e-05,
1879
+ "loss": 1.386,
1880
+ "step": 156000
1881
+ },
1882
+ {
1883
+ "epoch": 2.02,
1884
+ "learning_rate": 1.6368387573710182e-05,
1885
+ "loss": 1.3715,
1886
+ "step": 156500
1887
+ },
1888
+ {
1889
+ "epoch": 2.02,
1890
+ "learning_rate": 1.6260938332731617e-05,
1891
+ "loss": 1.3819,
1892
+ "step": 157000
1893
+ },
1894
+ {
1895
+ "epoch": 2.03,
1896
+ "learning_rate": 1.615348909175306e-05,
1897
+ "loss": 1.3813,
1898
+ "step": 157500
1899
+ },
1900
+ {
1901
+ "epoch": 2.04,
1902
+ "learning_rate": 1.6046039850774494e-05,
1903
+ "loss": 1.3724,
1904
+ "step": 158000
1905
+ },
1906
+ {
1907
+ "epoch": 2.04,
1908
+ "learning_rate": 1.5938590609795935e-05,
1909
+ "loss": 1.3903,
1910
+ "step": 158500
1911
+ },
1912
+ {
1913
+ "epoch": 2.05,
1914
+ "learning_rate": 1.583114136881737e-05,
1915
+ "loss": 1.3853,
1916
+ "step": 159000
1917
+ },
1918
+ {
1919
+ "epoch": 2.06,
1920
+ "learning_rate": 1.572369212783881e-05,
1921
+ "loss": 1.3844,
1922
+ "step": 159500
1923
+ },
1924
+ {
1925
+ "epoch": 2.06,
1926
+ "learning_rate": 1.5616242886860246e-05,
1927
+ "loss": 1.3785,
1928
+ "step": 160000
1929
+ },
1930
+ {
1931
+ "epoch": 2.07,
1932
+ "learning_rate": 1.5508793645881688e-05,
1933
+ "loss": 1.3848,
1934
+ "step": 160500
1935
+ },
1936
+ {
1937
+ "epoch": 2.08,
1938
+ "learning_rate": 1.5401344404903126e-05,
1939
+ "loss": 1.3859,
1940
+ "step": 161000
1941
+ },
1942
+ {
1943
+ "epoch": 2.08,
1944
+ "learning_rate": 1.5293895163924564e-05,
1945
+ "loss": 1.3781,
1946
+ "step": 161500
1947
+ },
1948
+ {
1949
+ "epoch": 2.09,
1950
+ "learning_rate": 1.5186445922946002e-05,
1951
+ "loss": 1.3878,
1952
+ "step": 162000
1953
+ },
1954
+ {
1955
+ "epoch": 2.1,
1956
+ "learning_rate": 1.5078996681967439e-05,
1957
+ "loss": 1.3973,
1958
+ "step": 162500
1959
+ },
1960
+ {
1961
+ "epoch": 2.1,
1962
+ "learning_rate": 1.4971547440988879e-05,
1963
+ "loss": 1.3938,
1964
+ "step": 163000
1965
+ },
1966
+ {
1967
+ "epoch": 2.11,
1968
+ "learning_rate": 1.4864098200010315e-05,
1969
+ "loss": 1.3807,
1970
+ "step": 163500
1971
+ },
1972
+ {
1973
+ "epoch": 2.11,
1974
+ "learning_rate": 1.4756648959031755e-05,
1975
+ "loss": 1.3834,
1976
+ "step": 164000
1977
+ },
1978
+ {
1979
+ "epoch": 2.12,
1980
+ "learning_rate": 1.4649199718053191e-05,
1981
+ "loss": 1.377,
1982
+ "step": 164500
1983
+ },
1984
+ {
1985
+ "epoch": 2.13,
1986
+ "learning_rate": 1.4541750477074631e-05,
1987
+ "loss": 1.3941,
1988
+ "step": 165000
1989
+ },
1990
+ {
1991
+ "epoch": 2.13,
1992
+ "learning_rate": 1.4434301236096068e-05,
1993
+ "loss": 1.3831,
1994
+ "step": 165500
1995
+ },
1996
+ {
1997
+ "epoch": 2.14,
1998
+ "learning_rate": 1.4326851995117508e-05,
1999
+ "loss": 1.3762,
2000
+ "step": 166000
2001
+ },
2002
+ {
2003
+ "epoch": 2.15,
2004
+ "learning_rate": 1.4219402754138944e-05,
2005
+ "loss": 1.3809,
2006
+ "step": 166500
2007
+ },
2008
+ {
2009
+ "epoch": 2.15,
2010
+ "learning_rate": 1.4111953513160384e-05,
2011
+ "loss": 1.3951,
2012
+ "step": 167000
2013
+ },
2014
+ {
2015
+ "epoch": 2.16,
2016
+ "learning_rate": 1.4004504272181824e-05,
2017
+ "loss": 1.3764,
2018
+ "step": 167500
2019
+ },
2020
+ {
2021
+ "epoch": 2.17,
2022
+ "learning_rate": 1.389705503120326e-05,
2023
+ "loss": 1.3742,
2024
+ "step": 168000
2025
+ },
2026
+ {
2027
+ "epoch": 2.17,
2028
+ "learning_rate": 1.37896057902247e-05,
2029
+ "loss": 1.3959,
2030
+ "step": 168500
2031
+ },
2032
+ {
2033
+ "epoch": 2.18,
2034
+ "learning_rate": 1.3682156549246137e-05,
2035
+ "loss": 1.4023,
2036
+ "step": 169000
2037
+ },
2038
+ {
2039
+ "epoch": 2.19,
2040
+ "learning_rate": 1.3574707308267577e-05,
2041
+ "loss": 1.3785,
2042
+ "step": 169500
2043
+ },
2044
+ {
2045
+ "epoch": 2.19,
2046
+ "learning_rate": 1.3467258067289013e-05,
2047
+ "loss": 1.3918,
2048
+ "step": 170000
2049
+ },
2050
+ {
2051
+ "epoch": 2.2,
2052
+ "learning_rate": 1.3359808826310453e-05,
2053
+ "loss": 1.385,
2054
+ "step": 170500
2055
+ },
2056
+ {
2057
+ "epoch": 2.2,
2058
+ "learning_rate": 1.325235958533189e-05,
2059
+ "loss": 1.3876,
2060
+ "step": 171000
2061
+ },
2062
+ {
2063
+ "epoch": 2.21,
2064
+ "learning_rate": 1.314491034435333e-05,
2065
+ "loss": 1.3787,
2066
+ "step": 171500
2067
+ },
2068
+ {
2069
+ "epoch": 2.22,
2070
+ "learning_rate": 1.3037461103374766e-05,
2071
+ "loss": 1.3825,
2072
+ "step": 172000
2073
+ },
2074
+ {
2075
+ "epoch": 2.22,
2076
+ "learning_rate": 1.2930011862396206e-05,
2077
+ "loss": 1.3753,
2078
+ "step": 172500
2079
+ },
2080
+ {
2081
+ "epoch": 2.23,
2082
+ "learning_rate": 1.2822562621417644e-05,
2083
+ "loss": 1.3799,
2084
+ "step": 173000
2085
+ },
2086
+ {
2087
+ "epoch": 2.24,
2088
+ "learning_rate": 1.271511338043908e-05,
2089
+ "loss": 1.3925,
2090
+ "step": 173500
2091
+ },
2092
+ {
2093
+ "epoch": 2.24,
2094
+ "learning_rate": 1.260766413946052e-05,
2095
+ "loss": 1.3963,
2096
+ "step": 174000
2097
+ },
2098
+ {
2099
+ "epoch": 2.25,
2100
+ "learning_rate": 1.2500214898481956e-05,
2101
+ "loss": 1.3809,
2102
+ "step": 174500
2103
+ },
2104
+ {
2105
+ "epoch": 2.26,
2106
+ "learning_rate": 1.2392765657503395e-05,
2107
+ "loss": 1.3765,
2108
+ "step": 175000
2109
+ },
2110
+ {
2111
+ "epoch": 2.26,
2112
+ "learning_rate": 1.2285316416524835e-05,
2113
+ "loss": 1.3743,
2114
+ "step": 175500
2115
+ },
2116
+ {
2117
+ "epoch": 2.27,
2118
+ "learning_rate": 1.2177867175546273e-05,
2119
+ "loss": 1.3804,
2120
+ "step": 176000
2121
+ },
2122
+ {
2123
+ "epoch": 2.28,
2124
+ "learning_rate": 1.207041793456771e-05,
2125
+ "loss": 1.3934,
2126
+ "step": 176500
2127
+ },
2128
+ {
2129
+ "epoch": 2.28,
2130
+ "learning_rate": 1.1962968693589149e-05,
2131
+ "loss": 1.3794,
2132
+ "step": 177000
2133
+ },
2134
+ {
2135
+ "epoch": 2.29,
2136
+ "learning_rate": 1.1855519452610587e-05,
2137
+ "loss": 1.3877,
2138
+ "step": 177500
2139
+ },
2140
+ {
2141
+ "epoch": 2.3,
2142
+ "learning_rate": 1.1748070211632025e-05,
2143
+ "loss": 1.3823,
2144
+ "step": 178000
2145
+ },
2146
+ {
2147
+ "epoch": 2.3,
2148
+ "learning_rate": 1.1640620970653464e-05,
2149
+ "loss": 1.3787,
2150
+ "step": 178500
2151
+ },
2152
+ {
2153
+ "epoch": 2.31,
2154
+ "learning_rate": 1.1533171729674902e-05,
2155
+ "loss": 1.384,
2156
+ "step": 179000
2157
+ },
2158
+ {
2159
+ "epoch": 2.31,
2160
+ "learning_rate": 1.142572248869634e-05,
2161
+ "loss": 1.3867,
2162
+ "step": 179500
2163
+ },
2164
+ {
2165
+ "epoch": 2.32,
2166
+ "learning_rate": 1.1318273247717778e-05,
2167
+ "loss": 1.3754,
2168
+ "step": 180000
2169
+ },
2170
+ {
2171
+ "epoch": 2.33,
2172
+ "learning_rate": 1.1210824006739216e-05,
2173
+ "loss": 1.3752,
2174
+ "step": 180500
2175
+ },
2176
+ {
2177
+ "epoch": 2.33,
2178
+ "learning_rate": 1.1103374765760654e-05,
2179
+ "loss": 1.3791,
2180
+ "step": 181000
2181
+ },
2182
+ {
2183
+ "epoch": 2.34,
2184
+ "learning_rate": 1.0995925524782094e-05,
2185
+ "loss": 1.3675,
2186
+ "step": 181500
2187
+ },
2188
+ {
2189
+ "epoch": 2.35,
2190
+ "learning_rate": 1.0888476283803532e-05,
2191
+ "loss": 1.376,
2192
+ "step": 182000
2193
+ },
2194
+ {
2195
+ "epoch": 2.35,
2196
+ "learning_rate": 1.078102704282497e-05,
2197
+ "loss": 1.3851,
2198
+ "step": 182500
2199
+ },
2200
+ {
2201
+ "epoch": 2.36,
2202
+ "learning_rate": 1.0673577801846409e-05,
2203
+ "loss": 1.3845,
2204
+ "step": 183000
2205
+ },
2206
+ {
2207
+ "epoch": 2.37,
2208
+ "learning_rate": 1.0566128560867847e-05,
2209
+ "loss": 1.3812,
2210
+ "step": 183500
2211
+ },
2212
+ {
2213
+ "epoch": 2.37,
2214
+ "learning_rate": 1.0458679319889285e-05,
2215
+ "loss": 1.382,
2216
+ "step": 184000
2217
+ },
2218
+ {
2219
+ "epoch": 2.38,
2220
+ "learning_rate": 1.0351230078910723e-05,
2221
+ "loss": 1.3685,
2222
+ "step": 184500
2223
+ },
2224
+ {
2225
+ "epoch": 2.39,
2226
+ "learning_rate": 1.0243780837932161e-05,
2227
+ "loss": 1.3882,
2228
+ "step": 185000
2229
+ },
2230
+ {
2231
+ "epoch": 2.39,
2232
+ "learning_rate": 1.01363315969536e-05,
2233
+ "loss": 1.3859,
2234
+ "step": 185500
2235
+ },
2236
+ {
2237
+ "epoch": 2.4,
2238
+ "learning_rate": 1.0028882355975038e-05,
2239
+ "loss": 1.4014,
2240
+ "step": 186000
2241
+ },
2242
+ {
2243
+ "epoch": 2.4,
2244
+ "learning_rate": 9.921433114996476e-06,
2245
+ "loss": 1.3826,
2246
+ "step": 186500
2247
+ },
2248
+ {
2249
+ "epoch": 2.41,
2250
+ "learning_rate": 9.813983874017914e-06,
2251
+ "loss": 1.3864,
2252
+ "step": 187000
2253
+ },
2254
+ {
2255
+ "epoch": 2.42,
2256
+ "learning_rate": 9.706534633039354e-06,
2257
+ "loss": 1.3771,
2258
+ "step": 187500
2259
+ },
2260
+ {
2261
+ "epoch": 2.42,
2262
+ "learning_rate": 9.599085392060792e-06,
2263
+ "loss": 1.3816,
2264
+ "step": 188000
2265
+ },
2266
+ {
2267
+ "epoch": 2.43,
2268
+ "learning_rate": 9.49163615108223e-06,
2269
+ "loss": 1.3885,
2270
+ "step": 188500
2271
+ },
2272
+ {
2273
+ "epoch": 2.44,
2274
+ "learning_rate": 9.384186910103668e-06,
2275
+ "loss": 1.3761,
2276
+ "step": 189000
2277
+ },
2278
+ {
2279
+ "epoch": 2.44,
2280
+ "learning_rate": 9.276737669125107e-06,
2281
+ "loss": 1.373,
2282
+ "step": 189500
2283
+ },
2284
+ {
2285
+ "epoch": 2.45,
2286
+ "learning_rate": 9.169288428146545e-06,
2287
+ "loss": 1.3909,
2288
+ "step": 190000
2289
+ },
2290
+ {
2291
+ "epoch": 2.46,
2292
+ "learning_rate": 9.061839187167983e-06,
2293
+ "loss": 1.3814,
2294
+ "step": 190500
2295
+ },
2296
+ {
2297
+ "epoch": 2.46,
2298
+ "learning_rate": 8.954389946189421e-06,
2299
+ "loss": 1.3709,
2300
+ "step": 191000
2301
+ },
2302
+ {
2303
+ "epoch": 2.47,
2304
+ "learning_rate": 8.84694070521086e-06,
2305
+ "loss": 1.3769,
2306
+ "step": 191500
2307
+ },
2308
+ {
2309
+ "epoch": 2.48,
2310
+ "learning_rate": 8.739491464232297e-06,
2311
+ "loss": 1.3883,
2312
+ "step": 192000
2313
+ },
2314
+ {
2315
+ "epoch": 2.48,
2316
+ "learning_rate": 8.632042223253736e-06,
2317
+ "loss": 1.378,
2318
+ "step": 192500
2319
+ },
2320
+ {
2321
+ "epoch": 2.49,
2322
+ "learning_rate": 8.524592982275174e-06,
2323
+ "loss": 1.3836,
2324
+ "step": 193000
2325
+ },
2326
+ {
2327
+ "epoch": 2.49,
2328
+ "learning_rate": 8.417143741296612e-06,
2329
+ "loss": 1.3639,
2330
+ "step": 193500
2331
+ },
2332
+ {
2333
+ "epoch": 2.5,
2334
+ "learning_rate": 8.30969450031805e-06,
2335
+ "loss": 1.3793,
2336
+ "step": 194000
2337
+ },
2338
+ {
2339
+ "epoch": 2.51,
2340
+ "learning_rate": 8.202245259339488e-06,
2341
+ "loss": 1.3775,
2342
+ "step": 194500
2343
+ },
2344
+ {
2345
+ "epoch": 2.51,
2346
+ "learning_rate": 8.094796018360926e-06,
2347
+ "loss": 1.3835,
2348
+ "step": 195000
2349
+ },
2350
+ {
2351
+ "epoch": 2.52,
2352
+ "learning_rate": 7.987346777382365e-06,
2353
+ "loss": 1.3726,
2354
+ "step": 195500
2355
+ },
2356
+ {
2357
+ "epoch": 2.53,
2358
+ "learning_rate": 7.879897536403803e-06,
2359
+ "loss": 1.3776,
2360
+ "step": 196000
2361
+ },
2362
+ {
2363
+ "epoch": 2.53,
2364
+ "learning_rate": 7.772448295425241e-06,
2365
+ "loss": 1.3799,
2366
+ "step": 196500
2367
+ },
2368
+ {
2369
+ "epoch": 2.54,
2370
+ "learning_rate": 7.664999054446679e-06,
2371
+ "loss": 1.3907,
2372
+ "step": 197000
2373
+ },
2374
+ {
2375
+ "epoch": 2.55,
2376
+ "learning_rate": 7.557549813468118e-06,
2377
+ "loss": 1.3682,
2378
+ "step": 197500
2379
+ },
2380
+ {
2381
+ "epoch": 2.55,
2382
+ "learning_rate": 7.450100572489556e-06,
2383
+ "loss": 1.3865,
2384
+ "step": 198000
2385
+ },
2386
+ {
2387
+ "epoch": 2.56,
2388
+ "learning_rate": 7.342651331510994e-06,
2389
+ "loss": 1.3762,
2390
+ "step": 198500
2391
+ },
2392
+ {
2393
+ "epoch": 2.57,
2394
+ "learning_rate": 7.2352020905324326e-06,
2395
+ "loss": 1.3853,
2396
+ "step": 199000
2397
+ },
2398
+ {
2399
+ "epoch": 2.57,
2400
+ "learning_rate": 7.127752849553872e-06,
2401
+ "loss": 1.3772,
2402
+ "step": 199500
2403
+ },
2404
+ {
2405
+ "epoch": 2.58,
2406
+ "learning_rate": 7.02030360857531e-06,
2407
+ "loss": 1.381,
2408
+ "step": 200000
2409
+ },
2410
+ {
2411
+ "epoch": 2.59,
2412
+ "learning_rate": 6.912854367596748e-06,
2413
+ "loss": 1.3643,
2414
+ "step": 200500
2415
+ },
2416
+ {
2417
+ "epoch": 2.59,
2418
+ "learning_rate": 6.805405126618186e-06,
2419
+ "loss": 1.3785,
2420
+ "step": 201000
2421
+ },
2422
+ {
2423
+ "epoch": 2.6,
2424
+ "learning_rate": 6.697955885639624e-06,
2425
+ "loss": 1.3776,
2426
+ "step": 201500
2427
+ },
2428
+ {
2429
+ "epoch": 2.6,
2430
+ "learning_rate": 6.5905066446610624e-06,
2431
+ "loss": 1.3755,
2432
+ "step": 202000
2433
+ },
2434
+ {
2435
+ "epoch": 2.61,
2436
+ "learning_rate": 6.483057403682501e-06,
2437
+ "loss": 1.3665,
2438
+ "step": 202500
2439
+ },
2440
+ {
2441
+ "epoch": 2.62,
2442
+ "learning_rate": 6.375608162703939e-06,
2443
+ "loss": 1.3702,
2444
+ "step": 203000
2445
+ },
2446
+ {
2447
+ "epoch": 2.62,
2448
+ "learning_rate": 6.268158921725377e-06,
2449
+ "loss": 1.3705,
2450
+ "step": 203500
2451
+ },
2452
+ {
2453
+ "epoch": 2.63,
2454
+ "learning_rate": 6.160709680746816e-06,
2455
+ "loss": 1.3729,
2456
+ "step": 204000
2457
+ },
2458
+ {
2459
+ "epoch": 2.64,
2460
+ "learning_rate": 6.053260439768254e-06,
2461
+ "loss": 1.379,
2462
+ "step": 204500
2463
+ },
2464
+ {
2465
+ "epoch": 2.64,
2466
+ "learning_rate": 5.945811198789692e-06,
2467
+ "loss": 1.3712,
2468
+ "step": 205000
2469
+ },
2470
+ {
2471
+ "epoch": 2.65,
2472
+ "learning_rate": 5.8383619578111304e-06,
2473
+ "loss": 1.3822,
2474
+ "step": 205500
2475
+ },
2476
+ {
2477
+ "epoch": 2.66,
2478
+ "learning_rate": 5.730912716832569e-06,
2479
+ "loss": 1.3616,
2480
+ "step": 206000
2481
+ },
2482
+ {
2483
+ "epoch": 2.66,
2484
+ "learning_rate": 5.623463475854007e-06,
2485
+ "loss": 1.3717,
2486
+ "step": 206500
2487
+ },
2488
+ {
2489
+ "epoch": 2.67,
2490
+ "learning_rate": 5.516014234875446e-06,
2491
+ "loss": 1.3622,
2492
+ "step": 207000
2493
+ },
2494
+ {
2495
+ "epoch": 2.68,
2496
+ "learning_rate": 5.408564993896884e-06,
2497
+ "loss": 1.3731,
2498
+ "step": 207500
2499
+ },
2500
+ {
2501
+ "epoch": 2.68,
2502
+ "learning_rate": 5.301115752918321e-06,
2503
+ "loss": 1.3697,
2504
+ "step": 208000
2505
+ },
2506
+ {
2507
+ "epoch": 2.69,
2508
+ "learning_rate": 5.1936665119397594e-06,
2509
+ "loss": 1.3755,
2510
+ "step": 208500
2511
+ },
2512
+ {
2513
+ "epoch": 2.69,
2514
+ "learning_rate": 5.086217270961198e-06,
2515
+ "loss": 1.3643,
2516
+ "step": 209000
2517
+ },
2518
+ {
2519
+ "epoch": 2.7,
2520
+ "learning_rate": 4.978768029982636e-06,
2521
+ "loss": 1.3628,
2522
+ "step": 209500
2523
+ },
2524
+ {
2525
+ "epoch": 2.71,
2526
+ "learning_rate": 4.871318789004075e-06,
2527
+ "loss": 1.3788,
2528
+ "step": 210000
2529
+ },
2530
+ {
2531
+ "epoch": 2.71,
2532
+ "learning_rate": 4.763869548025513e-06,
2533
+ "loss": 1.3818,
2534
+ "step": 210500
2535
+ },
2536
+ {
2537
+ "epoch": 2.72,
2538
+ "learning_rate": 4.656420307046951e-06,
2539
+ "loss": 1.364,
2540
+ "step": 211000
2541
+ },
2542
+ {
2543
+ "epoch": 2.73,
2544
+ "learning_rate": 4.548971066068389e-06,
2545
+ "loss": 1.3706,
2546
+ "step": 211500
2547
+ },
2548
+ {
2549
+ "epoch": 2.73,
2550
+ "learning_rate": 4.4415218250898275e-06,
2551
+ "loss": 1.3761,
2552
+ "step": 212000
2553
+ },
2554
+ {
2555
+ "epoch": 2.74,
2556
+ "learning_rate": 4.334072584111266e-06,
2557
+ "loss": 1.3917,
2558
+ "step": 212500
2559
+ },
2560
+ {
2561
+ "epoch": 2.75,
2562
+ "learning_rate": 4.226623343132705e-06,
2563
+ "loss": 1.3662,
2564
+ "step": 213000
2565
+ },
2566
+ {
2567
+ "epoch": 2.75,
2568
+ "learning_rate": 4.119174102154143e-06,
2569
+ "loss": 1.3621,
2570
+ "step": 213500
2571
+ },
2572
+ {
2573
+ "epoch": 2.76,
2574
+ "learning_rate": 4.011724861175581e-06,
2575
+ "loss": 1.3642,
2576
+ "step": 214000
2577
+ },
2578
+ {
2579
+ "epoch": 2.77,
2580
+ "learning_rate": 3.904275620197019e-06,
2581
+ "loss": 1.3702,
2582
+ "step": 214500
2583
+ },
2584
+ {
2585
+ "epoch": 2.77,
2586
+ "learning_rate": 3.7968263792184573e-06,
2587
+ "loss": 1.3658,
2588
+ "step": 215000
2589
+ },
2590
+ {
2591
+ "epoch": 2.78,
2592
+ "learning_rate": 3.6893771382398955e-06,
2593
+ "loss": 1.366,
2594
+ "step": 215500
2595
+ },
2596
+ {
2597
+ "epoch": 2.79,
2598
+ "learning_rate": 3.581927897261334e-06,
2599
+ "loss": 1.3607,
2600
+ "step": 216000
2601
+ },
2602
+ {
2603
+ "epoch": 2.79,
2604
+ "learning_rate": 3.4744786562827722e-06,
2605
+ "loss": 1.363,
2606
+ "step": 216500
2607
+ },
2608
+ {
2609
+ "epoch": 2.8,
2610
+ "learning_rate": 3.3670294153042104e-06,
2611
+ "loss": 1.3701,
2612
+ "step": 217000
2613
+ },
2614
+ {
2615
+ "epoch": 2.8,
2616
+ "learning_rate": 3.2595801743256486e-06,
2617
+ "loss": 1.3768,
2618
+ "step": 217500
2619
+ },
2620
+ {
2621
+ "epoch": 2.81,
2622
+ "learning_rate": 3.1521309333470867e-06,
2623
+ "loss": 1.3662,
2624
+ "step": 218000
2625
+ },
2626
+ {
2627
+ "epoch": 2.82,
2628
+ "learning_rate": 3.0446816923685253e-06,
2629
+ "loss": 1.3735,
2630
+ "step": 218500
2631
+ },
2632
+ {
2633
+ "epoch": 2.82,
2634
+ "learning_rate": 2.9372324513899635e-06,
2635
+ "loss": 1.3601,
2636
+ "step": 219000
2637
+ },
2638
+ {
2639
+ "epoch": 2.83,
2640
+ "learning_rate": 2.829783210411402e-06,
2641
+ "loss": 1.3665,
2642
+ "step": 219500
2643
+ },
2644
+ {
2645
+ "epoch": 2.84,
2646
+ "learning_rate": 2.7223339694328402e-06,
2647
+ "loss": 1.368,
2648
+ "step": 220000
2649
+ },
2650
+ {
2651
+ "epoch": 2.84,
2652
+ "learning_rate": 2.6148847284542784e-06,
2653
+ "loss": 1.3595,
2654
+ "step": 220500
2655
+ },
2656
+ {
2657
+ "epoch": 2.85,
2658
+ "learning_rate": 2.5074354874757166e-06,
2659
+ "loss": 1.3587,
2660
+ "step": 221000
2661
+ },
2662
+ {
2663
+ "epoch": 2.86,
2664
+ "learning_rate": 2.3999862464971547e-06,
2665
+ "loss": 1.3645,
2666
+ "step": 221500
2667
+ },
2668
+ {
2669
+ "epoch": 2.86,
2670
+ "learning_rate": 2.292537005518593e-06,
2671
+ "loss": 1.3652,
2672
+ "step": 222000
2673
+ },
2674
+ {
2675
+ "epoch": 2.87,
2676
+ "learning_rate": 2.1850877645400315e-06,
2677
+ "loss": 1.3677,
2678
+ "step": 222500
2679
+ },
2680
+ {
2681
+ "epoch": 2.88,
2682
+ "learning_rate": 2.0776385235614697e-06,
2683
+ "loss": 1.3684,
2684
+ "step": 223000
2685
+ },
2686
+ {
2687
+ "epoch": 2.88,
2688
+ "learning_rate": 1.970189282582908e-06,
2689
+ "loss": 1.3752,
2690
+ "step": 223500
2691
+ },
2692
+ {
2693
+ "epoch": 2.89,
2694
+ "learning_rate": 1.8627400416043464e-06,
2695
+ "loss": 1.3587,
2696
+ "step": 224000
2697
+ },
2698
+ {
2699
+ "epoch": 2.89,
2700
+ "learning_rate": 1.7552908006257844e-06,
2701
+ "loss": 1.3758,
2702
+ "step": 224500
2703
+ },
2704
+ {
2705
+ "epoch": 2.9,
2706
+ "learning_rate": 1.6478415596472225e-06,
2707
+ "loss": 1.3667,
2708
+ "step": 225000
2709
+ },
2710
+ {
2711
+ "epoch": 2.91,
2712
+ "learning_rate": 1.540392318668661e-06,
2713
+ "loss": 1.3651,
2714
+ "step": 225500
2715
+ },
2716
+ {
2717
+ "epoch": 2.91,
2718
+ "learning_rate": 1.4329430776900993e-06,
2719
+ "loss": 1.3721,
2720
+ "step": 226000
2721
+ },
2722
+ {
2723
+ "epoch": 2.92,
2724
+ "learning_rate": 1.3254938367115377e-06,
2725
+ "loss": 1.3664,
2726
+ "step": 226500
2727
+ },
2728
+ {
2729
+ "epoch": 2.93,
2730
+ "learning_rate": 1.2180445957329758e-06,
2731
+ "loss": 1.3687,
2732
+ "step": 227000
2733
+ },
2734
+ {
2735
+ "epoch": 2.93,
2736
+ "learning_rate": 1.110595354754414e-06,
2737
+ "loss": 1.3712,
2738
+ "step": 227500
2739
+ },
2740
+ {
2741
+ "epoch": 2.94,
2742
+ "learning_rate": 1.0031461137758524e-06,
2743
+ "loss": 1.3519,
2744
+ "step": 228000
2745
+ },
2746
+ {
2747
+ "epoch": 2.95,
2748
+ "learning_rate": 8.956968727972906e-07,
2749
+ "loss": 1.3697,
2750
+ "step": 228500
2751
+ },
2752
+ {
2753
+ "epoch": 2.95,
2754
+ "learning_rate": 7.882476318187288e-07,
2755
+ "loss": 1.3684,
2756
+ "step": 229000
2757
+ },
2758
+ {
2759
+ "epoch": 2.96,
2760
+ "learning_rate": 6.807983908401671e-07,
2761
+ "loss": 1.3664,
2762
+ "step": 229500
2763
+ },
2764
+ {
2765
+ "epoch": 2.97,
2766
+ "learning_rate": 5.733491498616055e-07,
2767
+ "loss": 1.3536,
2768
+ "step": 230000
2769
+ },
2770
+ {
2771
+ "epoch": 2.97,
2772
+ "learning_rate": 4.658999088830437e-07,
2773
+ "loss": 1.3731,
2774
+ "step": 230500
2775
+ },
2776
+ {
2777
+ "epoch": 2.98,
2778
+ "learning_rate": 3.5845066790448197e-07,
2779
+ "loss": 1.3634,
2780
+ "step": 231000
2781
+ },
2782
+ {
2783
+ "epoch": 2.98,
2784
+ "learning_rate": 2.510014269259202e-07,
2785
+ "loss": 1.3614,
2786
+ "step": 231500
2787
+ },
2788
+ {
2789
+ "epoch": 2.99,
2790
+ "learning_rate": 1.4355218594735846e-07,
2791
+ "loss": 1.3589,
2792
+ "step": 232000
2793
+ },
2794
+ {
2795
+ "epoch": 3.0,
2796
+ "learning_rate": 3.610294496879674e-08,
2797
+ "loss": 1.3624,
2798
+ "step": 232500
2799
+ },
2800
+ {
2801
+ "epoch": 3.0,
2802
+ "step": 232668,
2803
+ "total_flos": 8.28935012108206e+16,
2804
+ "train_loss": 1.506600116780431,
2805
+ "train_runtime": 70430.9377,
2806
+ "train_samples_per_second": 52.855,
2807
+ "train_steps_per_second": 3.303
2808
+ }
2809
+ ],
2810
+ "max_steps": 232668,
2811
+ "num_train_epochs": 3,
2812
+ "total_flos": 8.28935012108206e+16,
2813
+ "trial_name": null,
2814
+ "trial_params": null
2815
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:494d3e3e62d8f1a5648b63f14a7923104dbee545de5d963fe0398eb4ff1f3e7f
3
+ size 3247
vocab.json ADDED
The diff for this file is too large to render. See raw diff