Hugging Face's logo

YL95
/
mistral_naive_allFpML_v_proj_3080ti

Text Generation
TensorBoard
Trained with AutoTrain
Model card Files Metrics Community
YL95 commited on
Commit
46e1191
•
1 Parent(s): bed8abd

Upload folder using huggingface_hub

Browse files
Files changed (27) hide show
  1. README.md +9 -0
  2. adapter_config.json +22 -0
  3. adapter_model.bin +3 -0
  4. added_tokens.json +5 -0
  5. checkpoint-18018/README.md +34 -0
  6. checkpoint-18018/adapter_config.json +22 -0
  7. checkpoint-18018/adapter_model.bin +3 -0
  8. checkpoint-18018/added_tokens.json +5 -0
  9. checkpoint-18018/optimizer.pt +3 -0
  10. checkpoint-18018/rng_state.pth +3 -0
  11. checkpoint-18018/scheduler.pt +3 -0
  12. checkpoint-18018/special_tokens_map.json +11 -0
  13. checkpoint-18018/tokenizer.json +0 -0
  14. checkpoint-18018/tokenizer.model +3 -0
  15. checkpoint-18018/tokenizer_config.json +44 -0
  16. checkpoint-18018/trainer_state.json +2365 -0
  17. checkpoint-18018/training_args.bin +3 -0
  18. runs/Oct03_16-54-05_AM-3080TI-VR-PC/events.out.tfevents.1696348445.AM-3080TI-VR-PC.708.0 +3 -0
  19. runs/Oct03_16-56-07_AM-3080TI-VR-PC/events.out.tfevents.1696348567.AM-3080TI-VR-PC.769.0 +3 -0
  20. runs/Oct03_16-57-18_AM-3080TI-VR-PC/events.out.tfevents.1696348638.AM-3080TI-VR-PC.827.0 +3 -0
  21. runs/Oct03_16-58-27_AM-3080TI-VR-PC/events.out.tfevents.1696348708.AM-3080TI-VR-PC.884.0 +3 -0
  22. special_tokens_map.json +11 -0
  23. tokenizer.json +0 -0
  24. tokenizer.model +3 -0
  25. tokenizer_config.json +44 -0
  26. training_args.bin +3 -0
  27. training_params.json +1 -0
README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - autotrain
4
+ - text-generation
5
+ widget:
6
+ - text: "I love AutoTrain because "
7
+ ---
8
+
9
+ # Model Trained Using AutoTrain
adapter_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "bn22/Mistral-7B-Instruct-v0.1-sharded",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "lora_alpha": 32,
12
+ "lora_dropout": 0.05,
13
+ "modules_to_save": null,
14
+ "peft_type": "LORA",
15
+ "r": 16,
16
+ "rank_pattern": {},
17
+ "revision": null,
18
+ "target_modules": [
19
+ "v_proj"
20
+ ],
21
+ "task_type": "CAUSAL_LM"
22
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3b3937bece125325b96a0d892d1a1e9e8f5950e7f021768b11d7bf63c79530b
3
+ size 10508529
added_tokens.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "</s>": 2,
3
+ "<s>": 1,
4
+ "<unk>": 0
5
+ }
checkpoint-18018/README.md ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - quant_method: bitsandbytes
9
+ - load_in_8bit: False
10
+ - load_in_4bit: True
11
+ - llm_int8_threshold: 6.0
12
+ - llm_int8_skip_modules: None
13
+ - llm_int8_enable_fp32_cpu_offload: False
14
+ - llm_int8_has_fp16_weight: False
15
+ - bnb_4bit_quant_type: nf4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float16
18
+
19
+ The following `bitsandbytes` quantization config was used during training:
20
+ - quant_method: bitsandbytes
21
+ - load_in_8bit: False
22
+ - load_in_4bit: True
23
+ - llm_int8_threshold: 6.0
24
+ - llm_int8_skip_modules: None
25
+ - llm_int8_enable_fp32_cpu_offload: False
26
+ - llm_int8_has_fp16_weight: False
27
+ - bnb_4bit_quant_type: nf4
28
+ - bnb_4bit_use_double_quant: False
29
+ - bnb_4bit_compute_dtype: float16
30
+ ### Framework versions
31
+
32
+ - PEFT 0.6.0.dev0
33
+
34
+ - PEFT 0.6.0.dev0
checkpoint-18018/adapter_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "bn22/Mistral-7B-Instruct-v0.1-sharded",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "lora_alpha": 32,
12
+ "lora_dropout": 0.05,
13
+ "modules_to_save": null,
14
+ "peft_type": "LORA",
15
+ "r": 16,
16
+ "rank_pattern": {},
17
+ "revision": null,
18
+ "target_modules": [
19
+ "v_proj"
20
+ ],
21
+ "task_type": "CAUSAL_LM"
22
+ }
checkpoint-18018/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3b3937bece125325b96a0d892d1a1e9e8f5950e7f021768b11d7bf63c79530b
3
+ size 10508529
checkpoint-18018/added_tokens.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "</s>": 2,
3
+ "<s>": 1,
4
+ "<unk>": 0
5
+ }
checkpoint-18018/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2fce1b372dae54de58066d1c2a5c686705c162d18715ccd964bf76d2688d70d2
3
+ size 21025285
checkpoint-18018/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:19099753a4f37a899c08e9d6fa05add0b03f6b7b272e646f1af8bb60ceebf276
3
+ size 14575
checkpoint-18018/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84af0308eb9ceec8aa8a8e64fd1a4795e530d0144fdc43fcd4b46691a8faf332
3
+ size 627
checkpoint-18018/special_tokens_map.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<unk>",
4
+ "<s>",
5
+ "</s>"
6
+ ],
7
+ "bos_token": "<s>",
8
+ "eos_token": "</s>",
9
+ "pad_token": "</s>",
10
+ "unk_token": "<unk>"
11
+ }
checkpoint-18018/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-18018/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
checkpoint-18018/tokenizer_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<unk>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<s>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ }
27
+ },
28
+ "additional_special_tokens": [
29
+ "<unk>",
30
+ "<s>",
31
+ "</s>"
32
+ ],
33
+ "bos_token": "<s>",
34
+ "clean_up_tokenization_spaces": false,
35
+ "eos_token": "</s>",
36
+ "legacy": true,
37
+ "model_max_length": 1024,
38
+ "pad_token": "</s>",
39
+ "sp_model_kwargs": {},
40
+ "spaces_between_special_tokens": false,
41
+ "tokenizer_class": "LlamaTokenizer",
42
+ "unk_token": "<unk>",
43
+ "use_default_system_prompt": true
44
+ }
checkpoint-18018/trainer_state.json ADDED
@@ -0,0 +1,2365 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 24.414529914529915,
5
+ "eval_steps": 500,
6
+ "global_step": 18018,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.2,
13
+ "learning_rate": 2.5527192008879026e-06,
14
+ "loss": 0.8199,
15
+ "step": 46
16
+ },
17
+ {
18
+ "epoch": 0.39,
19
+ "learning_rate": 5.105438401775805e-06,
20
+ "loss": 0.8791,
21
+ "step": 92
22
+ },
23
+ {
24
+ "epoch": 0.59,
25
+ "learning_rate": 7.658157602663706e-06,
26
+ "loss": 0.7995,
27
+ "step": 138
28
+ },
29
+ {
30
+ "epoch": 0.79,
31
+ "learning_rate": 1.021087680355161e-05,
32
+ "loss": 0.7234,
33
+ "step": 184
34
+ },
35
+ {
36
+ "epoch": 0.98,
37
+ "learning_rate": 1.2763596004439513e-05,
38
+ "loss": 0.6077,
39
+ "step": 230
40
+ },
41
+ {
42
+ "epoch": 1.18,
43
+ "learning_rate": 1.5316315205327412e-05,
44
+ "loss": 0.5855,
45
+ "step": 276
46
+ },
47
+ {
48
+ "epoch": 1.38,
49
+ "learning_rate": 1.786903440621532e-05,
50
+ "loss": 0.5165,
51
+ "step": 322
52
+ },
53
+ {
54
+ "epoch": 1.57,
55
+ "learning_rate": 2.042175360710322e-05,
56
+ "loss": 0.5176,
57
+ "step": 368
58
+ },
59
+ {
60
+ "epoch": 1.77,
61
+ "learning_rate": 2.297447280799112e-05,
62
+ "loss": 0.4479,
63
+ "step": 414
64
+ },
65
+ {
66
+ "epoch": 1.97,
67
+ "learning_rate": 2.5527192008879026e-05,
68
+ "loss": 0.4451,
69
+ "step": 460
70
+ },
71
+ {
72
+ "epoch": 2.16,
73
+ "learning_rate": 2.807991120976693e-05,
74
+ "loss": 0.4133,
75
+ "step": 506
76
+ },
77
+ {
78
+ "epoch": 2.36,
79
+ "learning_rate": 3.0632630410654825e-05,
80
+ "loss": 0.3851,
81
+ "step": 552
82
+ },
83
+ {
84
+ "epoch": 2.56,
85
+ "learning_rate": 3.3185349611542734e-05,
86
+ "loss": 0.3973,
87
+ "step": 598
88
+ },
89
+ {
90
+ "epoch": 2.75,
91
+ "learning_rate": 3.573806881243064e-05,
92
+ "loss": 0.3546,
93
+ "step": 644
94
+ },
95
+ {
96
+ "epoch": 2.95,
97
+ "learning_rate": 3.829078801331853e-05,
98
+ "loss": 0.329,
99
+ "step": 690
100
+ },
101
+ {
102
+ "epoch": 3.15,
103
+ "learning_rate": 4.084350721420644e-05,
104
+ "loss": 0.3476,
105
+ "step": 736
106
+ },
107
+ {
108
+ "epoch": 1.06,
109
+ "learning_rate": 4.3396226415094345e-05,
110
+ "loss": 0.3039,
111
+ "step": 782
112
+ },
113
+ {
114
+ "epoch": 1.25,
115
+ "learning_rate": 4.594894561598224e-05,
116
+ "loss": 0.3156,
117
+ "step": 828
118
+ },
119
+ {
120
+ "epoch": 1.45,
121
+ "learning_rate": 4.850166481687014e-05,
122
+ "loss": 0.3129,
123
+ "step": 874
124
+ },
125
+ {
126
+ "epoch": 1.65,
127
+ "learning_rate": 5.105438401775805e-05,
128
+ "loss": 0.2968,
129
+ "step": 920
130
+ },
131
+ {
132
+ "epoch": 1.84,
133
+ "learning_rate": 5.360710321864595e-05,
134
+ "loss": 0.3168,
135
+ "step": 966
136
+ },
137
+ {
138
+ "epoch": 2.04,
139
+ "learning_rate": 5.615982241953386e-05,
140
+ "loss": 0.3347,
141
+ "step": 1012
142
+ },
143
+ {
144
+ "epoch": 2.24,
145
+ "learning_rate": 5.871254162042176e-05,
146
+ "loss": 0.3011,
147
+ "step": 1058
148
+ },
149
+ {
150
+ "epoch": 2.43,
151
+ "learning_rate": 6.126526082130965e-05,
152
+ "loss": 0.2856,
153
+ "step": 1104
154
+ },
155
+ {
156
+ "epoch": 2.63,
157
+ "learning_rate": 6.381798002219757e-05,
158
+ "loss": 0.2626,
159
+ "step": 1150
160
+ },
161
+ {
162
+ "epoch": 2.82,
163
+ "learning_rate": 6.637069922308547e-05,
164
+ "loss": 0.305,
165
+ "step": 1196
166
+ },
167
+ {
168
+ "epoch": 3.02,
169
+ "learning_rate": 6.892341842397336e-05,
170
+ "loss": 0.2895,
171
+ "step": 1242
172
+ },
173
+ {
174
+ "epoch": 3.22,
175
+ "learning_rate": 7.147613762486127e-05,
176
+ "loss": 0.2791,
177
+ "step": 1288
178
+ },
179
+ {
180
+ "epoch": 3.41,
181
+ "learning_rate": 7.402885682574918e-05,
182
+ "loss": 0.2542,
183
+ "step": 1334
184
+ },
185
+ {
186
+ "epoch": 3.61,
187
+ "learning_rate": 7.658157602663707e-05,
188
+ "loss": 0.2776,
189
+ "step": 1380
190
+ },
191
+ {
192
+ "epoch": 3.81,
193
+ "learning_rate": 7.913429522752498e-05,
194
+ "loss": 0.2573,
195
+ "step": 1426
196
+ },
197
+ {
198
+ "epoch": 4.0,
199
+ "learning_rate": 8.168701442841288e-05,
200
+ "loss": 0.2282,
201
+ "step": 1472
202
+ },
203
+ {
204
+ "epoch": 4.2,
205
+ "learning_rate": 8.423973362930077e-05,
206
+ "loss": 0.2645,
207
+ "step": 1518
208
+ },
209
+ {
210
+ "epoch": 2.11,
211
+ "learning_rate": 8.679245283018869e-05,
212
+ "loss": 0.2183,
213
+ "step": 1564
214
+ },
215
+ {
216
+ "epoch": 2.31,
217
+ "learning_rate": 8.934517203107659e-05,
218
+ "loss": 0.2505,
219
+ "step": 1610
220
+ },
221
+ {
222
+ "epoch": 2.5,
223
+ "learning_rate": 9.189789123196448e-05,
224
+ "loss": 0.2229,
225
+ "step": 1656
226
+ },
227
+ {
228
+ "epoch": 2.7,
229
+ "learning_rate": 9.44506104328524e-05,
230
+ "loss": 0.2404,
231
+ "step": 1702
232
+ },
233
+ {
234
+ "epoch": 2.9,
235
+ "learning_rate": 9.700332963374029e-05,
236
+ "loss": 0.2336,
237
+ "step": 1748
238
+ },
239
+ {
240
+ "epoch": 3.09,
241
+ "learning_rate": 9.955604883462819e-05,
242
+ "loss": 0.2178,
243
+ "step": 1794
244
+ },
245
+ {
246
+ "epoch": 3.29,
247
+ "learning_rate": 9.976566354218057e-05,
248
+ "loss": 0.1597,
249
+ "step": 1840
250
+ },
251
+ {
252
+ "epoch": 3.49,
253
+ "learning_rate": 9.948199309324125e-05,
254
+ "loss": 0.1837,
255
+ "step": 1886
256
+ },
257
+ {
258
+ "epoch": 3.68,
259
+ "learning_rate": 9.919832264430194e-05,
260
+ "loss": 0.2176,
261
+ "step": 1932
262
+ },
263
+ {
264
+ "epoch": 3.88,
265
+ "learning_rate": 9.891465219536261e-05,
266
+ "loss": 0.1992,
267
+ "step": 1978
268
+ },
269
+ {
270
+ "epoch": 4.08,
271
+ "learning_rate": 9.863098174642329e-05,
272
+ "loss": 0.1929,
273
+ "step": 2024
274
+ },
275
+ {
276
+ "epoch": 4.27,
277
+ "learning_rate": 9.834731129748396e-05,
278
+ "loss": 0.2017,
279
+ "step": 2070
280
+ },
281
+ {
282
+ "epoch": 4.47,
283
+ "learning_rate": 9.806364084854464e-05,
284
+ "loss": 0.1916,
285
+ "step": 2116
286
+ },
287
+ {
288
+ "epoch": 4.67,
289
+ "learning_rate": 9.777997039960533e-05,
290
+ "loss": 0.205,
291
+ "step": 2162
292
+ },
293
+ {
294
+ "epoch": 4.86,
295
+ "learning_rate": 9.749629995066602e-05,
296
+ "loss": 0.2272,
297
+ "step": 2208
298
+ },
299
+ {
300
+ "epoch": 5.06,
301
+ "learning_rate": 9.72126295017267e-05,
302
+ "loss": 0.2049,
303
+ "step": 2254
304
+ },
305
+ {
306
+ "epoch": 5.26,
307
+ "learning_rate": 9.692895905278737e-05,
308
+ "loss": 0.172,
309
+ "step": 2300
310
+ },
311
+ {
312
+ "epoch": 3.17,
313
+ "learning_rate": 9.664528860384806e-05,
314
+ "loss": 0.1737,
315
+ "step": 2346
316
+ },
317
+ {
318
+ "epoch": 3.36,
319
+ "learning_rate": 9.636161815490874e-05,
320
+ "loss": 0.1516,
321
+ "step": 2392
322
+ },
323
+ {
324
+ "epoch": 3.56,
325
+ "learning_rate": 9.607794770596941e-05,
326
+ "loss": 0.1686,
327
+ "step": 2438
328
+ },
329
+ {
330
+ "epoch": 3.76,
331
+ "learning_rate": 9.579427725703009e-05,
332
+ "loss": 0.1555,
333
+ "step": 2484
334
+ },
335
+ {
336
+ "epoch": 3.95,
337
+ "learning_rate": 9.551060680809078e-05,
338
+ "loss": 0.1519,
339
+ "step": 2530
340
+ },
341
+ {
342
+ "epoch": 4.15,
343
+ "learning_rate": 9.522693635915147e-05,
344
+ "loss": 0.1844,
345
+ "step": 2576
346
+ },
347
+ {
348
+ "epoch": 4.35,
349
+ "learning_rate": 9.494326591021215e-05,
350
+ "loss": 0.1493,
351
+ "step": 2622
352
+ },
353
+ {
354
+ "epoch": 4.54,
355
+ "learning_rate": 9.465959546127282e-05,
356
+ "loss": 0.155,
357
+ "step": 2668
358
+ },
359
+ {
360
+ "epoch": 4.74,
361
+ "learning_rate": 9.43759250123335e-05,
362
+ "loss": 0.153,
363
+ "step": 2714
364
+ },
365
+ {
366
+ "epoch": 4.94,
367
+ "learning_rate": 9.409225456339417e-05,
368
+ "loss": 0.157,
369
+ "step": 2760
370
+ },
371
+ {
372
+ "epoch": 5.13,
373
+ "learning_rate": 9.380858411445486e-05,
374
+ "loss": 0.1439,
375
+ "step": 2806
376
+ },
377
+ {
378
+ "epoch": 5.33,
379
+ "learning_rate": 9.352491366551554e-05,
380
+ "loss": 0.1429,
381
+ "step": 2852
382
+ },
383
+ {
384
+ "epoch": 5.53,
385
+ "learning_rate": 9.324124321657623e-05,
386
+ "loss": 0.1609,
387
+ "step": 2898
388
+ },
389
+ {
390
+ "epoch": 5.72,
391
+ "learning_rate": 9.29575727676369e-05,
392
+ "loss": 0.1664,
393
+ "step": 2944
394
+ },
395
+ {
396
+ "epoch": 5.92,
397
+ "learning_rate": 9.26739023186976e-05,
398
+ "loss": 0.1534,
399
+ "step": 2990
400
+ },
401
+ {
402
+ "epoch": 6.12,
403
+ "learning_rate": 9.239023186975827e-05,
404
+ "loss": 0.1518,
405
+ "step": 3036
406
+ },
407
+ {
408
+ "epoch": 4.03,
409
+ "learning_rate": 9.210656142081895e-05,
410
+ "loss": 0.1435,
411
+ "step": 3082
412
+ },
413
+ {
414
+ "epoch": 4.22,
415
+ "learning_rate": 9.182289097187962e-05,
416
+ "loss": 0.1335,
417
+ "step": 3128
418
+ },
419
+ {
420
+ "epoch": 4.42,
421
+ "learning_rate": 9.15392205229403e-05,
422
+ "loss": 0.1231,
423
+ "step": 3174
424
+ },
425
+ {
426
+ "epoch": 4.62,
427
+ "learning_rate": 9.125555007400099e-05,
428
+ "loss": 0.1237,
429
+ "step": 3220
430
+ },
431
+ {
432
+ "epoch": 4.81,
433
+ "learning_rate": 9.097187962506168e-05,
434
+ "loss": 0.1312,
435
+ "step": 3266
436
+ },
437
+ {
438
+ "epoch": 5.01,
439
+ "learning_rate": 9.068820917612236e-05,
440
+ "loss": 0.133,
441
+ "step": 3312
442
+ },
443
+ {
444
+ "epoch": 5.21,
445
+ "learning_rate": 9.040453872718303e-05,
446
+ "loss": 0.1324,
447
+ "step": 3358
448
+ },
449
+ {
450
+ "epoch": 5.4,
451
+ "learning_rate": 9.012086827824372e-05,
452
+ "loss": 0.1232,
453
+ "step": 3404
454
+ },
455
+ {
456
+ "epoch": 5.6,
457
+ "learning_rate": 8.98371978293044e-05,
458
+ "loss": 0.1104,
459
+ "step": 3450
460
+ },
461
+ {
462
+ "epoch": 5.79,
463
+ "learning_rate": 8.955352738036507e-05,
464
+ "loss": 0.1166,
465
+ "step": 3496
466
+ },
467
+ {
468
+ "epoch": 5.99,
469
+ "learning_rate": 8.926985693142575e-05,
470
+ "loss": 0.1114,
471
+ "step": 3542
472
+ },
473
+ {
474
+ "epoch": 6.19,
475
+ "learning_rate": 8.898618648248643e-05,
476
+ "loss": 0.1276,
477
+ "step": 3588
478
+ },
479
+ {
480
+ "epoch": 6.38,
481
+ "learning_rate": 8.870251603354712e-05,
482
+ "loss": 0.127,
483
+ "step": 3634
484
+ },
485
+ {
486
+ "epoch": 6.58,
487
+ "learning_rate": 8.84188455846078e-05,
488
+ "loss": 0.1255,
489
+ "step": 3680
490
+ },
491
+ {
492
+ "epoch": 6.78,
493
+ "learning_rate": 8.813517513566848e-05,
494
+ "loss": 0.1269,
495
+ "step": 3726
496
+ },
497
+ {
498
+ "epoch": 6.97,
499
+ "learning_rate": 8.785150468672916e-05,
500
+ "loss": 0.1339,
501
+ "step": 3772
502
+ },
503
+ {
504
+ "epoch": 7.17,
505
+ "learning_rate": 8.756783423778983e-05,
506
+ "loss": 0.1326,
507
+ "step": 3818
508
+ },
509
+ {
510
+ "epoch": 5.08,
511
+ "learning_rate": 8.728416378885052e-05,
512
+ "loss": 0.1203,
513
+ "step": 3864
514
+ },
515
+ {
516
+ "epoch": 5.28,
517
+ "learning_rate": 8.70004933399112e-05,
518
+ "loss": 0.1026,
519
+ "step": 3910
520
+ },
521
+ {
522
+ "epoch": 5.47,
523
+ "learning_rate": 8.671682289097188e-05,
524
+ "loss": 0.1252,
525
+ "step": 3956
526
+ },
527
+ {
528
+ "epoch": 5.67,
529
+ "learning_rate": 8.643315244203257e-05,
530
+ "loss": 0.1052,
531
+ "step": 4002
532
+ },
533
+ {
534
+ "epoch": 5.87,
535
+ "learning_rate": 8.614948199309326e-05,
536
+ "loss": 0.1188,
537
+ "step": 4048
538
+ },
539
+ {
540
+ "epoch": 6.06,
541
+ "learning_rate": 8.586581154415393e-05,
542
+ "loss": 0.1113,
543
+ "step": 4094
544
+ },
545
+ {
546
+ "epoch": 6.26,
547
+ "learning_rate": 8.558214109521461e-05,
548
+ "loss": 0.0934,
549
+ "step": 4140
550
+ },
551
+ {
552
+ "epoch": 6.46,
553
+ "learning_rate": 8.529847064627528e-05,
554
+ "loss": 0.1168,
555
+ "step": 4186
556
+ },
557
+ {
558
+ "epoch": 6.65,
559
+ "learning_rate": 8.501480019733596e-05,
560
+ "loss": 0.1113,
561
+ "step": 4232
562
+ },
563
+ {
564
+ "epoch": 6.85,
565
+ "learning_rate": 8.473112974839665e-05,
566
+ "loss": 0.0944,
567
+ "step": 4278
568
+ },
569
+ {
570
+ "epoch": 7.05,
571
+ "learning_rate": 8.444745929945733e-05,
572
+ "loss": 0.0956,
573
+ "step": 4324
574
+ },
575
+ {
576
+ "epoch": 7.24,
577
+ "learning_rate": 8.416378885051802e-05,
578
+ "loss": 0.092,
579
+ "step": 4370
580
+ },
581
+ {
582
+ "epoch": 7.44,
583
+ "learning_rate": 8.388011840157869e-05,
584
+ "loss": 0.0899,
585
+ "step": 4416
586
+ },
587
+ {
588
+ "epoch": 7.64,
589
+ "learning_rate": 8.359644795263937e-05,
590
+ "loss": 0.111,
591
+ "step": 4462
592
+ },
593
+ {
594
+ "epoch": 7.83,
595
+ "learning_rate": 8.331277750370006e-05,
596
+ "loss": 0.114,
597
+ "step": 4508
598
+ },
599
+ {
600
+ "epoch": 8.03,
601
+ "learning_rate": 8.302910705476073e-05,
602
+ "loss": 0.0906,
603
+ "step": 4554
604
+ },
605
+ {
606
+ "epoch": 8.23,
607
+ "learning_rate": 8.274543660582141e-05,
608
+ "loss": 0.1053,
609
+ "step": 4600
610
+ },
611
+ {
612
+ "epoch": 6.14,
613
+ "learning_rate": 8.246176615688209e-05,
614
+ "loss": 0.0983,
615
+ "step": 4646
616
+ },
617
+ {
618
+ "epoch": 6.33,
619
+ "learning_rate": 8.217809570794278e-05,
620
+ "loss": 0.0864,
621
+ "step": 4692
622
+ },
623
+ {
624
+ "epoch": 6.53,
625
+ "learning_rate": 8.189442525900347e-05,
626
+ "loss": 0.0857,
627
+ "step": 4738
628
+ },
629
+ {
630
+ "epoch": 6.73,
631
+ "learning_rate": 8.161075481006414e-05,
632
+ "loss": 0.0979,
633
+ "step": 4784
634
+ },
635
+ {
636
+ "epoch": 6.92,
637
+ "learning_rate": 8.132708436112482e-05,
638
+ "loss": 0.0919,
639
+ "step": 4830
640
+ },
641
+ {
642
+ "epoch": 7.12,
643
+ "learning_rate": 8.10434139121855e-05,
644
+ "loss": 0.0758,
645
+ "step": 4876
646
+ },
647
+ {
648
+ "epoch": 7.32,
649
+ "learning_rate": 8.075974346324618e-05,
650
+ "loss": 0.0948,
651
+ "step": 4922
652
+ },
653
+ {
654
+ "epoch": 7.51,
655
+ "learning_rate": 8.047607301430686e-05,
656
+ "loss": 0.0856,
657
+ "step": 4968
658
+ },
659
+ {
660
+ "epoch": 7.71,
661
+ "learning_rate": 8.019240256536754e-05,
662
+ "loss": 0.0677,
663
+ "step": 5014
664
+ },
665
+ {
666
+ "epoch": 7.91,
667
+ "learning_rate": 7.990873211642823e-05,
668
+ "loss": 0.0855,
669
+ "step": 5060
670
+ },
671
+ {
672
+ "epoch": 8.1,
673
+ "learning_rate": 7.96250616674889e-05,
674
+ "loss": 0.0896,
675
+ "step": 5106
676
+ },
677
+ {
678
+ "epoch": 8.3,
679
+ "learning_rate": 7.934139121854959e-05,
680
+ "loss": 0.0999,
681
+ "step": 5152
682
+ },
683
+ {
684
+ "epoch": 8.5,
685
+ "learning_rate": 7.905772076961027e-05,
686
+ "loss": 0.0866,
687
+ "step": 5198
688
+ },
689
+ {
690
+ "epoch": 8.69,
691
+ "learning_rate": 7.877405032067094e-05,
692
+ "loss": 0.0821,
693
+ "step": 5244
694
+ },
695
+ {
696
+ "epoch": 8.89,
697
+ "learning_rate": 7.849037987173162e-05,
698
+ "loss": 0.0826,
699
+ "step": 5290
700
+ },
701
+ {
702
+ "epoch": 9.09,
703
+ "learning_rate": 7.820670942279231e-05,
704
+ "loss": 0.0872,
705
+ "step": 5336
706
+ },
707
+ {
708
+ "epoch": 9.28,
709
+ "learning_rate": 7.792303897385299e-05,
710
+ "loss": 0.1113,
711
+ "step": 5382
712
+ },
713
+ {
714
+ "epoch": 7.19,
715
+ "learning_rate": 7.763936852491366e-05,
716
+ "loss": 0.0628,
717
+ "step": 5428
718
+ },
719
+ {
720
+ "epoch": 7.39,
721
+ "learning_rate": 7.735569807597435e-05,
722
+ "loss": 0.0747,
723
+ "step": 5474
724
+ },
725
+ {
726
+ "epoch": 7.59,
727
+ "learning_rate": 7.707202762703503e-05,
728
+ "loss": 0.0653,
729
+ "step": 5520
730
+ },
731
+ {
732
+ "epoch": 7.78,
733
+ "learning_rate": 7.678835717809572e-05,
734
+ "loss": 0.0852,
735
+ "step": 5566
736
+ },
737
+ {
738
+ "epoch": 7.98,
739
+ "learning_rate": 7.65046867291564e-05,
740
+ "loss": 0.0661,
741
+ "step": 5612
742
+ },
743
+ {
744
+ "epoch": 8.18,
745
+ "learning_rate": 7.622101628021707e-05,
746
+ "loss": 0.0683,
747
+ "step": 5658
748
+ },
749
+ {
750
+ "epoch": 8.37,
751
+ "learning_rate": 7.593734583127775e-05,
752
+ "loss": 0.0865,
753
+ "step": 5704
754
+ },
755
+ {
756
+ "epoch": 8.57,
757
+ "learning_rate": 7.565367538233844e-05,
758
+ "loss": 0.0722,
759
+ "step": 5750
760
+ },
761
+ {
762
+ "epoch": 8.76,
763
+ "learning_rate": 7.537000493339911e-05,
764
+ "loss": 0.0802,
765
+ "step": 5796
766
+ },
767
+ {
768
+ "epoch": 8.96,
769
+ "learning_rate": 7.50863344844598e-05,
770
+ "loss": 0.0693,
771
+ "step": 5842
772
+ },
773
+ {
774
+ "epoch": 9.16,
775
+ "learning_rate": 7.480266403552048e-05,
776
+ "loss": 0.0817,
777
+ "step": 5888
778
+ },
779
+ {
780
+ "epoch": 9.35,
781
+ "learning_rate": 7.451899358658115e-05,
782
+ "loss": 0.0844,
783
+ "step": 5934
784
+ },
785
+ {
786
+ "epoch": 9.55,
787
+ "learning_rate": 7.423532313764184e-05,
788
+ "loss": 0.0803,
789
+ "step": 5980
790
+ },
791
+ {
792
+ "epoch": 9.75,
793
+ "learning_rate": 7.395165268870252e-05,
794
+ "loss": 0.0815,
795
+ "step": 6026
796
+ },
797
+ {
798
+ "epoch": 9.94,
799
+ "learning_rate": 7.36679822397632e-05,
800
+ "loss": 0.0871,
801
+ "step": 6072
802
+ },
803
+ {
804
+ "epoch": 10.14,
805
+ "learning_rate": 7.338431179082387e-05,
806
+ "loss": 0.0763,
807
+ "step": 6118
808
+ },
809
+ {
810
+ "epoch": 8.05,
811
+ "learning_rate": 7.310064134188456e-05,
812
+ "loss": 0.065,
813
+ "step": 6164
814
+ },
815
+ {
816
+ "epoch": 8.25,
817
+ "learning_rate": 7.281697089294525e-05,
818
+ "loss": 0.0629,
819
+ "step": 6210
820
+ },
821
+ {
822
+ "epoch": 8.44,
823
+ "learning_rate": 7.253330044400593e-05,
824
+ "loss": 0.0712,
825
+ "step": 6256
826
+ },
827
+ {
828
+ "epoch": 8.64,
829
+ "learning_rate": 7.22496299950666e-05,
830
+ "loss": 0.0645,
831
+ "step": 6302
832
+ },
833
+ {
834
+ "epoch": 8.84,
835
+ "learning_rate": 7.196595954612728e-05,
836
+ "loss": 0.0725,
837
+ "step": 6348
838
+ },
839
+ {
840
+ "epoch": 9.03,
841
+ "learning_rate": 7.168228909718797e-05,
842
+ "loss": 0.0573,
843
+ "step": 6394
844
+ },
845
+ {
846
+ "epoch": 9.23,
847
+ "learning_rate": 7.139861864824865e-05,
848
+ "loss": 0.0621,
849
+ "step": 6440
850
+ },
851
+ {
852
+ "epoch": 9.43,
853
+ "learning_rate": 7.111494819930932e-05,
854
+ "loss": 0.06,
855
+ "step": 6486
856
+ },
857
+ {
858
+ "epoch": 9.62,
859
+ "learning_rate": 7.083127775037001e-05,
860
+ "loss": 0.0646,
861
+ "step": 6532
862
+ },
863
+ {
864
+ "epoch": 9.82,
865
+ "learning_rate": 7.054760730143069e-05,
866
+ "loss": 0.0575,
867
+ "step": 6578
868
+ },
869
+ {
870
+ "epoch": 10.02,
871
+ "learning_rate": 7.026393685249138e-05,
872
+ "loss": 0.0634,
873
+ "step": 6624
874
+ },
875
+ {
876
+ "epoch": 10.21,
877
+ "learning_rate": 6.998026640355205e-05,
878
+ "loss": 0.0568,
879
+ "step": 6670
880
+ },
881
+ {
882
+ "epoch": 10.41,
883
+ "learning_rate": 6.969659595461273e-05,
884
+ "loss": 0.0652,
885
+ "step": 6716
886
+ },
887
+ {
888
+ "epoch": 10.61,
889
+ "learning_rate": 6.941292550567341e-05,
890
+ "loss": 0.0795,
891
+ "step": 6762
892
+ },
893
+ {
894
+ "epoch": 10.8,
895
+ "learning_rate": 6.912925505673408e-05,
896
+ "loss": 0.064,
897
+ "step": 6808
898
+ },
899
+ {
900
+ "epoch": 11.0,
901
+ "learning_rate": 6.884558460779477e-05,
902
+ "loss": 0.0846,
903
+ "step": 6854
904
+ },
905
+ {
906
+ "epoch": 11.2,
907
+ "learning_rate": 6.856191415885546e-05,
908
+ "loss": 0.0587,
909
+ "step": 6900
910
+ },
911
+ {
912
+ "epoch": 9.11,
913
+ "learning_rate": 6.827824370991614e-05,
914
+ "loss": 0.0663,
915
+ "step": 6946
916
+ },
917
+ {
918
+ "epoch": 9.3,
919
+ "learning_rate": 6.799457326097681e-05,
920
+ "loss": 0.0534,
921
+ "step": 6992
922
+ },
923
+ {
924
+ "epoch": 9.5,
925
+ "learning_rate": 6.77109028120375e-05,
926
+ "loss": 0.0466,
927
+ "step": 7038
928
+ },
929
+ {
930
+ "epoch": 9.7,
931
+ "learning_rate": 6.742723236309818e-05,
932
+ "loss": 0.0529,
933
+ "step": 7084
934
+ },
935
+ {
936
+ "epoch": 9.89,
937
+ "learning_rate": 6.714356191415886e-05,
938
+ "loss": 0.0595,
939
+ "step": 7130
940
+ },
941
+ {
942
+ "epoch": 10.09,
943
+ "learning_rate": 6.685989146521953e-05,
944
+ "loss": 0.0547,
945
+ "step": 7176
946
+ },
947
+ {
948
+ "epoch": 10.29,
949
+ "learning_rate": 6.657622101628021e-05,
950
+ "loss": 0.0571,
951
+ "step": 7222
952
+ },
953
+ {
954
+ "epoch": 10.48,
955
+ "learning_rate": 6.62925505673409e-05,
956
+ "loss": 0.0601,
957
+ "step": 7268
958
+ },
959
+ {
960
+ "epoch": 10.68,
961
+ "learning_rate": 6.600888011840159e-05,
962
+ "loss": 0.0508,
963
+ "step": 7314
964
+ },
965
+ {
966
+ "epoch": 10.88,
967
+ "learning_rate": 6.572520966946226e-05,
968
+ "loss": 0.0572,
969
+ "step": 7360
970
+ },
971
+ {
972
+ "epoch": 11.07,
973
+ "learning_rate": 6.544153922052294e-05,
974
+ "loss": 0.0591,
975
+ "step": 7406
976
+ },
977
+ {
978
+ "epoch": 11.27,
979
+ "learning_rate": 6.515786877158362e-05,
980
+ "loss": 0.0708,
981
+ "step": 7452
982
+ },
983
+ {
984
+ "epoch": 11.47,
985
+ "learning_rate": 6.487419832264431e-05,
986
+ "loss": 0.0711,
987
+ "step": 7498
988
+ },
989
+ {
990
+ "epoch": 11.66,
991
+ "learning_rate": 6.459052787370498e-05,
992
+ "loss": 0.0514,
993
+ "step": 7544
994
+ },
995
+ {
996
+ "epoch": 11.86,
997
+ "learning_rate": 6.430685742476566e-05,
998
+ "loss": 0.0607,
999
+ "step": 7590
1000
+ },
1001
+ {
1002
+ "epoch": 12.06,
1003
+ "learning_rate": 6.402318697582635e-05,
1004
+ "loss": 0.0491,
1005
+ "step": 7636
1006
+ },
1007
+ {
1008
+ "epoch": 12.25,
1009
+ "learning_rate": 6.373951652688704e-05,
1010
+ "loss": 0.0577,
1011
+ "step": 7682
1012
+ },
1013
+ {
1014
+ "epoch": 10.16,
1015
+ "learning_rate": 6.345584607794771e-05,
1016
+ "loss": 0.0416,
1017
+ "step": 7728
1018
+ },
1019
+ {
1020
+ "epoch": 10.36,
1021
+ "learning_rate": 6.317217562900839e-05,
1022
+ "loss": 0.0572,
1023
+ "step": 7774
1024
+ },
1025
+ {
1026
+ "epoch": 10.56,
1027
+ "learning_rate": 6.288850518006907e-05,
1028
+ "loss": 0.0554,
1029
+ "step": 7820
1030
+ },
1031
+ {
1032
+ "epoch": 10.75,
1033
+ "learning_rate": 6.260483473112974e-05,
1034
+ "loss": 0.0457,
1035
+ "step": 7866
1036
+ },
1037
+ {
1038
+ "epoch": 10.95,
1039
+ "learning_rate": 6.232116428219043e-05,
1040
+ "loss": 0.0528,
1041
+ "step": 7912
1042
+ },
1043
+ {
1044
+ "epoch": 11.15,
1045
+ "learning_rate": 6.203749383325111e-05,
1046
+ "loss": 0.0524,
1047
+ "step": 7958
1048
+ },
1049
+ {
1050
+ "epoch": 11.34,
1051
+ "learning_rate": 6.17538233843118e-05,
1052
+ "loss": 0.0631,
1053
+ "step": 8004
1054
+ },
1055
+ {
1056
+ "epoch": 11.54,
1057
+ "learning_rate": 6.147015293537247e-05,
1058
+ "loss": 0.042,
1059
+ "step": 8050
1060
+ },
1061
+ {
1062
+ "epoch": 11.74,
1063
+ "learning_rate": 6.118648248643315e-05,
1064
+ "loss": 0.0422,
1065
+ "step": 8096
1066
+ },
1067
+ {
1068
+ "epoch": 11.93,
1069
+ "learning_rate": 6.090281203749384e-05,
1070
+ "loss": 0.0511,
1071
+ "step": 8142
1072
+ },
1073
+ {
1074
+ "epoch": 12.13,
1075
+ "learning_rate": 6.061914158855452e-05,
1076
+ "loss": 0.0526,
1077
+ "step": 8188
1078
+ },
1079
+ {
1080
+ "epoch": 12.32,
1081
+ "learning_rate": 6.033547113961519e-05,
1082
+ "loss": 0.0454,
1083
+ "step": 8234
1084
+ },
1085
+ {
1086
+ "epoch": 12.52,
1087
+ "learning_rate": 6.0051800690675876e-05,
1088
+ "loss": 0.0462,
1089
+ "step": 8280
1090
+ },
1091
+ {
1092
+ "epoch": 12.72,
1093
+ "learning_rate": 5.9768130241736566e-05,
1094
+ "loss": 0.0499,
1095
+ "step": 8326
1096
+ },
1097
+ {
1098
+ "epoch": 12.91,
1099
+ "learning_rate": 5.948445979279724e-05,
1100
+ "loss": 0.0444,
1101
+ "step": 8372
1102
+ },
1103
+ {
1104
+ "epoch": 13.11,
1105
+ "learning_rate": 5.920078934385792e-05,
1106
+ "loss": 0.0462,
1107
+ "step": 8418
1108
+ },
1109
+ {
1110
+ "epoch": 11.02,
1111
+ "learning_rate": 5.89171188949186e-05,
1112
+ "loss": 0.0443,
1113
+ "step": 8464
1114
+ },
1115
+ {
1116
+ "epoch": 11.22,
1117
+ "learning_rate": 5.863344844597928e-05,
1118
+ "loss": 0.0394,
1119
+ "step": 8510
1120
+ },
1121
+ {
1122
+ "epoch": 11.41,
1123
+ "learning_rate": 5.834977799703997e-05,
1124
+ "loss": 0.0463,
1125
+ "step": 8556
1126
+ },
1127
+ {
1128
+ "epoch": 11.61,
1129
+ "learning_rate": 5.806610754810064e-05,
1130
+ "loss": 0.0396,
1131
+ "step": 8602
1132
+ },
1133
+ {
1134
+ "epoch": 11.81,
1135
+ "learning_rate": 5.7782437099161326e-05,
1136
+ "loss": 0.0409,
1137
+ "step": 8648
1138
+ },
1139
+ {
1140
+ "epoch": 12.0,
1141
+ "learning_rate": 5.7498766650222e-05,
1142
+ "loss": 0.0407,
1143
+ "step": 8694
1144
+ },
1145
+ {
1146
+ "epoch": 12.2,
1147
+ "learning_rate": 5.721509620128269e-05,
1148
+ "loss": 0.0442,
1149
+ "step": 8740
1150
+ },
1151
+ {
1152
+ "epoch": 12.4,
1153
+ "learning_rate": 5.693142575234337e-05,
1154
+ "loss": 0.0437,
1155
+ "step": 8786
1156
+ },
1157
+ {
1158
+ "epoch": 12.59,
1159
+ "learning_rate": 5.664775530340405e-05,
1160
+ "loss": 0.0368,
1161
+ "step": 8832
1162
+ },
1163
+ {
1164
+ "epoch": 12.79,
1165
+ "learning_rate": 5.636408485446473e-05,
1166
+ "loss": 0.0418,
1167
+ "step": 8878
1168
+ },
1169
+ {
1170
+ "epoch": 12.99,
1171
+ "learning_rate": 5.60804144055254e-05,
1172
+ "loss": 0.0412,
1173
+ "step": 8924
1174
+ },
1175
+ {
1176
+ "epoch": 13.18,
1177
+ "learning_rate": 5.579674395658609e-05,
1178
+ "loss": 0.0385,
1179
+ "step": 8970
1180
+ },
1181
+ {
1182
+ "epoch": 13.38,
1183
+ "learning_rate": 5.5513073507646776e-05,
1184
+ "loss": 0.0469,
1185
+ "step": 9016
1186
+ },
1187
+ {
1188
+ "epoch": 13.58,
1189
+ "learning_rate": 5.522940305870745e-05,
1190
+ "loss": 0.0402,
1191
+ "step": 9062
1192
+ },
1193
+ {
1194
+ "epoch": 13.77,
1195
+ "learning_rate": 5.494573260976813e-05,
1196
+ "loss": 0.0514,
1197
+ "step": 9108
1198
+ },
1199
+ {
1200
+ "epoch": 13.97,
1201
+ "learning_rate": 5.466206216082881e-05,
1202
+ "loss": 0.045,
1203
+ "step": 9154
1204
+ },
1205
+ {
1206
+ "epoch": 14.17,
1207
+ "learning_rate": 5.43783917118895e-05,
1208
+ "loss": 0.0523,
1209
+ "step": 9200
1210
+ },
1211
+ {
1212
+ "epoch": 12.08,
1213
+ "learning_rate": 5.409472126295018e-05,
1214
+ "loss": 0.0338,
1215
+ "step": 9246
1216
+ },
1217
+ {
1218
+ "epoch": 12.27,
1219
+ "learning_rate": 5.381105081401085e-05,
1220
+ "loss": 0.0358,
1221
+ "step": 9292
1222
+ },
1223
+ {
1224
+ "epoch": 12.47,
1225
+ "learning_rate": 5.3527380365071536e-05,
1226
+ "loss": 0.0387,
1227
+ "step": 9338
1228
+ },
1229
+ {
1230
+ "epoch": 12.67,
1231
+ "learning_rate": 5.3243709916132226e-05,
1232
+ "loss": 0.0289,
1233
+ "step": 9384
1234
+ },
1235
+ {
1236
+ "epoch": 12.86,
1237
+ "learning_rate": 5.29600394671929e-05,
1238
+ "loss": 0.0396,
1239
+ "step": 9430
1240
+ },
1241
+ {
1242
+ "epoch": 13.06,
1243
+ "learning_rate": 5.267636901825358e-05,
1244
+ "loss": 0.0426,
1245
+ "step": 9476
1246
+ },
1247
+ {
1248
+ "epoch": 13.26,
1249
+ "learning_rate": 5.239269856931426e-05,
1250
+ "loss": 0.0345,
1251
+ "step": 9522
1252
+ },
1253
+ {
1254
+ "epoch": 13.45,
1255
+ "learning_rate": 5.210902812037494e-05,
1256
+ "loss": 0.0394,
1257
+ "step": 9568
1258
+ },
1259
+ {
1260
+ "epoch": 13.65,
1261
+ "learning_rate": 5.182535767143563e-05,
1262
+ "loss": 0.0349,
1263
+ "step": 9614
1264
+ },
1265
+ {
1266
+ "epoch": 13.85,
1267
+ "learning_rate": 5.15416872224963e-05,
1268
+ "loss": 0.0322,
1269
+ "step": 9660
1270
+ },
1271
+ {
1272
+ "epoch": 14.04,
1273
+ "learning_rate": 5.1258016773556986e-05,
1274
+ "loss": 0.0372,
1275
+ "step": 9706
1276
+ },
1277
+ {
1278
+ "epoch": 14.24,
1279
+ "learning_rate": 5.097434632461766e-05,
1280
+ "loss": 0.0334,
1281
+ "step": 9752
1282
+ },
1283
+ {
1284
+ "epoch": 14.44,
1285
+ "learning_rate": 5.069067587567834e-05,
1286
+ "loss": 0.0375,
1287
+ "step": 9798
1288
+ },
1289
+ {
1290
+ "epoch": 14.63,
1291
+ "learning_rate": 5.040700542673903e-05,
1292
+ "loss": 0.0396,
1293
+ "step": 9844
1294
+ },
1295
+ {
1296
+ "epoch": 14.83,
1297
+ "learning_rate": 5.0123334977799704e-05,
1298
+ "loss": 0.0356,
1299
+ "step": 9890
1300
+ },
1301
+ {
1302
+ "epoch": 15.03,
1303
+ "learning_rate": 4.983966452886039e-05,
1304
+ "loss": 0.0458,
1305
+ "step": 9936
1306
+ },
1307
+ {
1308
+ "epoch": 15.22,
1309
+ "learning_rate": 4.955599407992107e-05,
1310
+ "loss": 0.036,
1311
+ "step": 9982
1312
+ },
1313
+ {
1314
+ "epoch": 13.13,
1315
+ "learning_rate": 4.9272323630981746e-05,
1316
+ "loss": 0.028,
1317
+ "step": 10028
1318
+ },
1319
+ {
1320
+ "epoch": 13.33,
1321
+ "learning_rate": 4.898865318204243e-05,
1322
+ "loss": 0.0229,
1323
+ "step": 10074
1324
+ },
1325
+ {
1326
+ "epoch": 13.53,
1327
+ "learning_rate": 4.870498273310311e-05,
1328
+ "loss": 0.0353,
1329
+ "step": 10120
1330
+ },
1331
+ {
1332
+ "epoch": 13.72,
1333
+ "learning_rate": 4.842131228416379e-05,
1334
+ "loss": 0.0278,
1335
+ "step": 10166
1336
+ },
1337
+ {
1338
+ "epoch": 13.92,
1339
+ "learning_rate": 4.813764183522447e-05,
1340
+ "loss": 0.0296,
1341
+ "step": 10212
1342
+ },
1343
+ {
1344
+ "epoch": 14.12,
1345
+ "learning_rate": 4.7853971386285154e-05,
1346
+ "loss": 0.0284,
1347
+ "step": 10258
1348
+ },
1349
+ {
1350
+ "epoch": 14.31,
1351
+ "learning_rate": 4.757030093734584e-05,
1352
+ "loss": 0.0299,
1353
+ "step": 10304
1354
+ },
1355
+ {
1356
+ "epoch": 14.51,
1357
+ "learning_rate": 4.728663048840651e-05,
1358
+ "loss": 0.0309,
1359
+ "step": 10350
1360
+ },
1361
+ {
1362
+ "epoch": 14.71,
1363
+ "learning_rate": 4.7002960039467196e-05,
1364
+ "loss": 0.0283,
1365
+ "step": 10396
1366
+ },
1367
+ {
1368
+ "epoch": 14.9,
1369
+ "learning_rate": 4.671928959052788e-05,
1370
+ "loss": 0.0349,
1371
+ "step": 10442
1372
+ },
1373
+ {
1374
+ "epoch": 15.1,
1375
+ "learning_rate": 4.6435619141588555e-05,
1376
+ "loss": 0.0343,
1377
+ "step": 10488
1378
+ },
1379
+ {
1380
+ "epoch": 15.29,
1381
+ "learning_rate": 4.615194869264924e-05,
1382
+ "loss": 0.0347,
1383
+ "step": 10534
1384
+ },
1385
+ {
1386
+ "epoch": 15.49,
1387
+ "learning_rate": 4.5868278243709914e-05,
1388
+ "loss": 0.035,
1389
+ "step": 10580
1390
+ },
1391
+ {
1392
+ "epoch": 15.69,
1393
+ "learning_rate": 4.55846077947706e-05,
1394
+ "loss": 0.0284,
1395
+ "step": 10626
1396
+ },
1397
+ {
1398
+ "epoch": 15.88,
1399
+ "learning_rate": 4.530093734583128e-05,
1400
+ "loss": 0.0408,
1401
+ "step": 10672
1402
+ },
1403
+ {
1404
+ "epoch": 16.08,
1405
+ "learning_rate": 4.501726689689196e-05,
1406
+ "loss": 0.0386,
1407
+ "step": 10718
1408
+ },
1409
+ {
1410
+ "epoch": 16.28,
1411
+ "learning_rate": 4.473359644795264e-05,
1412
+ "loss": 0.0288,
1413
+ "step": 10764
1414
+ },
1415
+ {
1416
+ "epoch": 14.19,
1417
+ "learning_rate": 4.444992599901332e-05,
1418
+ "loss": 0.0274,
1419
+ "step": 10810
1420
+ },
1421
+ {
1422
+ "epoch": 14.38,
1423
+ "learning_rate": 4.4166255550074005e-05,
1424
+ "loss": 0.0268,
1425
+ "step": 10856
1426
+ },
1427
+ {
1428
+ "epoch": 14.58,
1429
+ "learning_rate": 4.388258510113468e-05,
1430
+ "loss": 0.0206,
1431
+ "step": 10902
1432
+ },
1433
+ {
1434
+ "epoch": 14.78,
1435
+ "learning_rate": 4.3598914652195364e-05,
1436
+ "loss": 0.0281,
1437
+ "step": 10948
1438
+ },
1439
+ {
1440
+ "epoch": 14.97,
1441
+ "learning_rate": 4.331524420325605e-05,
1442
+ "loss": 0.0242,
1443
+ "step": 10994
1444
+ },
1445
+ {
1446
+ "epoch": 15.17,
1447
+ "learning_rate": 4.303157375431673e-05,
1448
+ "loss": 0.0246,
1449
+ "step": 11040
1450
+ },
1451
+ {
1452
+ "epoch": 15.37,
1453
+ "learning_rate": 4.2747903305377406e-05,
1454
+ "loss": 0.0275,
1455
+ "step": 11086
1456
+ },
1457
+ {
1458
+ "epoch": 15.56,
1459
+ "learning_rate": 4.246423285643808e-05,
1460
+ "loss": 0.0259,
1461
+ "step": 11132
1462
+ },
1463
+ {
1464
+ "epoch": 15.76,
1465
+ "learning_rate": 4.218056240749877e-05,
1466
+ "loss": 0.0251,
1467
+ "step": 11178
1468
+ },
1469
+ {
1470
+ "epoch": 15.96,
1471
+ "learning_rate": 4.189689195855945e-05,
1472
+ "loss": 0.0229,
1473
+ "step": 11224
1474
+ },
1475
+ {
1476
+ "epoch": 16.15,
1477
+ "learning_rate": 4.161322150962013e-05,
1478
+ "loss": 0.0257,
1479
+ "step": 11270
1480
+ },
1481
+ {
1482
+ "epoch": 16.35,
1483
+ "learning_rate": 4.132955106068081e-05,
1484
+ "loss": 0.0269,
1485
+ "step": 11316
1486
+ },
1487
+ {
1488
+ "epoch": 16.55,
1489
+ "learning_rate": 4.10458806117415e-05,
1490
+ "loss": 0.0344,
1491
+ "step": 11362
1492
+ },
1493
+ {
1494
+ "epoch": 16.74,
1495
+ "learning_rate": 4.0762210162802173e-05,
1496
+ "loss": 0.0288,
1497
+ "step": 11408
1498
+ },
1499
+ {
1500
+ "epoch": 16.94,
1501
+ "learning_rate": 4.047853971386285e-05,
1502
+ "loss": 0.0287,
1503
+ "step": 11454
1504
+ },
1505
+ {
1506
+ "epoch": 17.14,
1507
+ "learning_rate": 4.019486926492353e-05,
1508
+ "loss": 0.0315,
1509
+ "step": 11500
1510
+ },
1511
+ {
1512
+ "epoch": 15.05,
1513
+ "learning_rate": 3.9911198815984215e-05,
1514
+ "loss": 0.0317,
1515
+ "step": 11546
1516
+ },
1517
+ {
1518
+ "epoch": 15.24,
1519
+ "learning_rate": 3.96275283670449e-05,
1520
+ "loss": 0.02,
1521
+ "step": 11592
1522
+ },
1523
+ {
1524
+ "epoch": 15.44,
1525
+ "learning_rate": 3.9343857918105575e-05,
1526
+ "loss": 0.0215,
1527
+ "step": 11638
1528
+ },
1529
+ {
1530
+ "epoch": 15.64,
1531
+ "learning_rate": 3.906018746916626e-05,
1532
+ "loss": 0.0276,
1533
+ "step": 11684
1534
+ },
1535
+ {
1536
+ "epoch": 15.83,
1537
+ "learning_rate": 3.877651702022694e-05,
1538
+ "loss": 0.0207,
1539
+ "step": 11730
1540
+ },
1541
+ {
1542
+ "epoch": 16.03,
1543
+ "learning_rate": 3.8492846571287617e-05,
1544
+ "loss": 0.0217,
1545
+ "step": 11776
1546
+ },
1547
+ {
1548
+ "epoch": 16.23,
1549
+ "learning_rate": 3.82091761223483e-05,
1550
+ "loss": 0.0254,
1551
+ "step": 11822
1552
+ },
1553
+ {
1554
+ "epoch": 16.42,
1555
+ "learning_rate": 3.7925505673408976e-05,
1556
+ "loss": 0.0245,
1557
+ "step": 11868
1558
+ },
1559
+ {
1560
+ "epoch": 16.62,
1561
+ "learning_rate": 3.7641835224469665e-05,
1562
+ "loss": 0.0275,
1563
+ "step": 11914
1564
+ },
1565
+ {
1566
+ "epoch": 16.82,
1567
+ "learning_rate": 3.735816477553034e-05,
1568
+ "loss": 0.0261,
1569
+ "step": 11960
1570
+ },
1571
+ {
1572
+ "epoch": 17.01,
1573
+ "learning_rate": 3.7074494326591025e-05,
1574
+ "loss": 0.0186,
1575
+ "step": 12006
1576
+ },
1577
+ {
1578
+ "epoch": 17.21,
1579
+ "learning_rate": 3.67908238776517e-05,
1580
+ "loss": 0.0186,
1581
+ "step": 12052
1582
+ },
1583
+ {
1584
+ "epoch": 17.41,
1585
+ "learning_rate": 3.650715342871239e-05,
1586
+ "loss": 0.0221,
1587
+ "step": 12098
1588
+ },
1589
+ {
1590
+ "epoch": 17.6,
1591
+ "learning_rate": 3.6223482979773067e-05,
1592
+ "loss": 0.0199,
1593
+ "step": 12144
1594
+ },
1595
+ {
1596
+ "epoch": 17.8,
1597
+ "learning_rate": 3.593981253083374e-05,
1598
+ "loss": 0.0241,
1599
+ "step": 12190
1600
+ },
1601
+ {
1602
+ "epoch": 18.0,
1603
+ "learning_rate": 3.5656142081894426e-05,
1604
+ "loss": 0.0224,
1605
+ "step": 12236
1606
+ },
1607
+ {
1608
+ "epoch": 18.19,
1609
+ "learning_rate": 3.537247163295511e-05,
1610
+ "loss": 0.0236,
1611
+ "step": 12282
1612
+ },
1613
+ {
1614
+ "epoch": 16.1,
1615
+ "learning_rate": 3.508880118401579e-05,
1616
+ "loss": 0.0206,
1617
+ "step": 12328
1618
+ },
1619
+ {
1620
+ "epoch": 16.3,
1621
+ "learning_rate": 3.480513073507647e-05,
1622
+ "loss": 0.0167,
1623
+ "step": 12374
1624
+ },
1625
+ {
1626
+ "epoch": 16.5,
1627
+ "learning_rate": 3.452146028613715e-05,
1628
+ "loss": 0.0209,
1629
+ "step": 12420
1630
+ },
1631
+ {
1632
+ "epoch": 16.69,
1633
+ "learning_rate": 3.4237789837197834e-05,
1634
+ "loss": 0.0188,
1635
+ "step": 12466
1636
+ },
1637
+ {
1638
+ "epoch": 16.89,
1639
+ "learning_rate": 3.395411938825851e-05,
1640
+ "loss": 0.0189,
1641
+ "step": 12512
1642
+ },
1643
+ {
1644
+ "epoch": 17.09,
1645
+ "learning_rate": 3.367044893931919e-05,
1646
+ "loss": 0.0229,
1647
+ "step": 12558
1648
+ },
1649
+ {
1650
+ "epoch": 17.28,
1651
+ "learning_rate": 3.338677849037987e-05,
1652
+ "loss": 0.0186,
1653
+ "step": 12604
1654
+ },
1655
+ {
1656
+ "epoch": 17.48,
1657
+ "learning_rate": 3.310310804144056e-05,
1658
+ "loss": 0.0218,
1659
+ "step": 12650
1660
+ },
1661
+ {
1662
+ "epoch": 17.68,
1663
+ "learning_rate": 3.2819437592501235e-05,
1664
+ "loss": 0.0165,
1665
+ "step": 12696
1666
+ },
1667
+ {
1668
+ "epoch": 17.87,
1669
+ "learning_rate": 3.253576714356192e-05,
1670
+ "loss": 0.0175,
1671
+ "step": 12742
1672
+ },
1673
+ {
1674
+ "epoch": 18.07,
1675
+ "learning_rate": 3.2252096694622594e-05,
1676
+ "loss": 0.0159,
1677
+ "step": 12788
1678
+ },
1679
+ {
1680
+ "epoch": 18.26,
1681
+ "learning_rate": 3.196842624568328e-05,
1682
+ "loss": 0.0174,
1683
+ "step": 12834
1684
+ },
1685
+ {
1686
+ "epoch": 18.46,
1687
+ "learning_rate": 3.168475579674396e-05,
1688
+ "loss": 0.0195,
1689
+ "step": 12880
1690
+ },
1691
+ {
1692
+ "epoch": 18.66,
1693
+ "learning_rate": 3.1401085347804636e-05,
1694
+ "loss": 0.0203,
1695
+ "step": 12926
1696
+ },
1697
+ {
1698
+ "epoch": 18.85,
1699
+ "learning_rate": 3.111741489886532e-05,
1700
+ "loss": 0.019,
1701
+ "step": 12972
1702
+ },
1703
+ {
1704
+ "epoch": 19.05,
1705
+ "learning_rate": 3.0833744449926e-05,
1706
+ "loss": 0.0197,
1707
+ "step": 13018
1708
+ },
1709
+ {
1710
+ "epoch": 19.25,
1711
+ "learning_rate": 3.0550074000986685e-05,
1712
+ "loss": 0.0221,
1713
+ "step": 13064
1714
+ },
1715
+ {
1716
+ "epoch": 17.16,
1717
+ "learning_rate": 3.026640355204736e-05,
1718
+ "loss": 0.0191,
1719
+ "step": 13110
1720
+ },
1721
+ {
1722
+ "epoch": 17.35,
1723
+ "learning_rate": 2.998273310310804e-05,
1724
+ "loss": 0.0138,
1725
+ "step": 13156
1726
+ },
1727
+ {
1728
+ "epoch": 17.55,
1729
+ "learning_rate": 2.9699062654168723e-05,
1730
+ "loss": 0.0149,
1731
+ "step": 13202
1732
+ },
1733
+ {
1734
+ "epoch": 17.75,
1735
+ "learning_rate": 2.9415392205229403e-05,
1736
+ "loss": 0.018,
1737
+ "step": 13248
1738
+ },
1739
+ {
1740
+ "epoch": 17.94,
1741
+ "learning_rate": 2.9131721756290086e-05,
1742
+ "loss": 0.0162,
1743
+ "step": 13294
1744
+ },
1745
+ {
1746
+ "epoch": 18.14,
1747
+ "learning_rate": 2.8848051307350765e-05,
1748
+ "loss": 0.0153,
1749
+ "step": 13340
1750
+ },
1751
+ {
1752
+ "epoch": 18.34,
1753
+ "learning_rate": 2.8564380858411448e-05,
1754
+ "loss": 0.0141,
1755
+ "step": 13386
1756
+ },
1757
+ {
1758
+ "epoch": 18.53,
1759
+ "learning_rate": 2.8280710409472128e-05,
1760
+ "loss": 0.013,
1761
+ "step": 13432
1762
+ },
1763
+ {
1764
+ "epoch": 18.73,
1765
+ "learning_rate": 2.7997039960532807e-05,
1766
+ "loss": 0.0184,
1767
+ "step": 13478
1768
+ },
1769
+ {
1770
+ "epoch": 18.93,
1771
+ "learning_rate": 2.771336951159349e-05,
1772
+ "loss": 0.0149,
1773
+ "step": 13524
1774
+ },
1775
+ {
1776
+ "epoch": 19.12,
1777
+ "learning_rate": 2.742969906265417e-05,
1778
+ "loss": 0.0159,
1779
+ "step": 13570
1780
+ },
1781
+ {
1782
+ "epoch": 19.32,
1783
+ "learning_rate": 2.7146028613714853e-05,
1784
+ "loss": 0.0171,
1785
+ "step": 13616
1786
+ },
1787
+ {
1788
+ "epoch": 19.52,
1789
+ "learning_rate": 2.6862358164775532e-05,
1790
+ "loss": 0.0174,
1791
+ "step": 13662
1792
+ },
1793
+ {
1794
+ "epoch": 19.71,
1795
+ "learning_rate": 2.6578687715836215e-05,
1796
+ "loss": 0.0159,
1797
+ "step": 13708
1798
+ },
1799
+ {
1800
+ "epoch": 19.91,
1801
+ "learning_rate": 2.629501726689689e-05,
1802
+ "loss": 0.0202,
1803
+ "step": 13754
1804
+ },
1805
+ {
1806
+ "epoch": 20.11,
1807
+ "learning_rate": 2.601134681795757e-05,
1808
+ "loss": 0.0182,
1809
+ "step": 13800
1810
+ },
1811
+ {
1812
+ "epoch": 18.02,
1813
+ "learning_rate": 2.5727676369018254e-05,
1814
+ "loss": 0.0138,
1815
+ "step": 13846
1816
+ },
1817
+ {
1818
+ "epoch": 18.21,
1819
+ "learning_rate": 2.5444005920078933e-05,
1820
+ "loss": 0.0136,
1821
+ "step": 13892
1822
+ },
1823
+ {
1824
+ "epoch": 18.41,
1825
+ "learning_rate": 2.5160335471139616e-05,
1826
+ "loss": 0.0135,
1827
+ "step": 13938
1828
+ },
1829
+ {
1830
+ "epoch": 18.61,
1831
+ "learning_rate": 2.4876665022200296e-05,
1832
+ "loss": 0.0125,
1833
+ "step": 13984
1834
+ },
1835
+ {
1836
+ "epoch": 18.8,
1837
+ "learning_rate": 2.459299457326098e-05,
1838
+ "loss": 0.014,
1839
+ "step": 14030
1840
+ },
1841
+ {
1842
+ "epoch": 19.0,
1843
+ "learning_rate": 2.430932412432166e-05,
1844
+ "loss": 0.0137,
1845
+ "step": 14076
1846
+ },
1847
+ {
1848
+ "epoch": 19.2,
1849
+ "learning_rate": 2.402565367538234e-05,
1850
+ "loss": 0.0162,
1851
+ "step": 14122
1852
+ },
1853
+ {
1854
+ "epoch": 19.39,
1855
+ "learning_rate": 2.374198322644302e-05,
1856
+ "loss": 0.0134,
1857
+ "step": 14168
1858
+ },
1859
+ {
1860
+ "epoch": 19.59,
1861
+ "learning_rate": 2.34583127775037e-05,
1862
+ "loss": 0.0113,
1863
+ "step": 14214
1864
+ },
1865
+ {
1866
+ "epoch": 19.79,
1867
+ "learning_rate": 2.3174642328564383e-05,
1868
+ "loss": 0.0134,
1869
+ "step": 14260
1870
+ },
1871
+ {
1872
+ "epoch": 19.98,
1873
+ "learning_rate": 2.2890971879625063e-05,
1874
+ "loss": 0.0135,
1875
+ "step": 14306
1876
+ },
1877
+ {
1878
+ "epoch": 20.18,
1879
+ "learning_rate": 2.2607301430685742e-05,
1880
+ "loss": 0.0132,
1881
+ "step": 14352
1882
+ },
1883
+ {
1884
+ "epoch": 20.38,
1885
+ "learning_rate": 2.2323630981746425e-05,
1886
+ "loss": 0.015,
1887
+ "step": 14398
1888
+ },
1889
+ {
1890
+ "epoch": 20.57,
1891
+ "learning_rate": 2.2039960532807105e-05,
1892
+ "loss": 0.0117,
1893
+ "step": 14444
1894
+ },
1895
+ {
1896
+ "epoch": 20.77,
1897
+ "learning_rate": 2.1756290083867788e-05,
1898
+ "loss": 0.0148,
1899
+ "step": 14490
1900
+ },
1901
+ {
1902
+ "epoch": 20.97,
1903
+ "learning_rate": 2.1472619634928467e-05,
1904
+ "loss": 0.0168,
1905
+ "step": 14536
1906
+ },
1907
+ {
1908
+ "epoch": 21.16,
1909
+ "learning_rate": 2.118894918598915e-05,
1910
+ "loss": 0.0139,
1911
+ "step": 14582
1912
+ },
1913
+ {
1914
+ "epoch": 19.07,
1915
+ "learning_rate": 2.0905278737049827e-05,
1916
+ "loss": 0.0112,
1917
+ "step": 14628
1918
+ },
1919
+ {
1920
+ "epoch": 19.27,
1921
+ "learning_rate": 2.062160828811051e-05,
1922
+ "loss": 0.0094,
1923
+ "step": 14674
1924
+ },
1925
+ {
1926
+ "epoch": 19.47,
1927
+ "learning_rate": 2.033793783917119e-05,
1928
+ "loss": 0.0092,
1929
+ "step": 14720
1930
+ },
1931
+ {
1932
+ "epoch": 19.66,
1933
+ "learning_rate": 2.0054267390231872e-05,
1934
+ "loss": 0.0111,
1935
+ "step": 14766
1936
+ },
1937
+ {
1938
+ "epoch": 19.86,
1939
+ "learning_rate": 1.977059694129255e-05,
1940
+ "loss": 0.0112,
1941
+ "step": 14812
1942
+ },
1943
+ {
1944
+ "epoch": 20.06,
1945
+ "learning_rate": 1.9486926492353234e-05,
1946
+ "loss": 0.0127,
1947
+ "step": 14858
1948
+ },
1949
+ {
1950
+ "epoch": 20.25,
1951
+ "learning_rate": 1.9203256043413914e-05,
1952
+ "loss": 0.0105,
1953
+ "step": 14904
1954
+ },
1955
+ {
1956
+ "epoch": 20.45,
1957
+ "learning_rate": 1.8919585594474594e-05,
1958
+ "loss": 0.0142,
1959
+ "step": 14950
1960
+ },
1961
+ {
1962
+ "epoch": 20.65,
1963
+ "learning_rate": 1.8635915145535273e-05,
1964
+ "loss": 0.0106,
1965
+ "step": 14996
1966
+ },
1967
+ {
1968
+ "epoch": 20.84,
1969
+ "learning_rate": 1.8352244696595956e-05,
1970
+ "loss": 0.0115,
1971
+ "step": 15042
1972
+ },
1973
+ {
1974
+ "epoch": 21.04,
1975
+ "learning_rate": 1.8068574247656636e-05,
1976
+ "loss": 0.0114,
1977
+ "step": 15088
1978
+ },
1979
+ {
1980
+ "epoch": 21.24,
1981
+ "learning_rate": 1.778490379871732e-05,
1982
+ "loss": 0.012,
1983
+ "step": 15134
1984
+ },
1985
+ {
1986
+ "epoch": 21.43,
1987
+ "learning_rate": 1.7501233349777998e-05,
1988
+ "loss": 0.0111,
1989
+ "step": 15180
1990
+ },
1991
+ {
1992
+ "epoch": 21.63,
1993
+ "learning_rate": 1.721756290083868e-05,
1994
+ "loss": 0.0127,
1995
+ "step": 15226
1996
+ },
1997
+ {
1998
+ "epoch": 21.82,
1999
+ "learning_rate": 1.693389245189936e-05,
2000
+ "loss": 0.0108,
2001
+ "step": 15272
2002
+ },
2003
+ {
2004
+ "epoch": 22.02,
2005
+ "learning_rate": 1.665022200296004e-05,
2006
+ "loss": 0.0123,
2007
+ "step": 15318
2008
+ },
2009
+ {
2010
+ "epoch": 22.22,
2011
+ "learning_rate": 1.636655155402072e-05,
2012
+ "loss": 0.0116,
2013
+ "step": 15364
2014
+ },
2015
+ {
2016
+ "epoch": 20.13,
2017
+ "learning_rate": 1.6082881105081403e-05,
2018
+ "loss": 0.0115,
2019
+ "step": 15410
2020
+ },
2021
+ {
2022
+ "epoch": 20.32,
2023
+ "learning_rate": 1.5799210656142082e-05,
2024
+ "loss": 0.0083,
2025
+ "step": 15456
2026
+ },
2027
+ {
2028
+ "epoch": 20.52,
2029
+ "learning_rate": 1.5515540207202765e-05,
2030
+ "loss": 0.0094,
2031
+ "step": 15502
2032
+ },
2033
+ {
2034
+ "epoch": 20.72,
2035
+ "learning_rate": 1.5231869758263445e-05,
2036
+ "loss": 0.0087,
2037
+ "step": 15548
2038
+ },
2039
+ {
2040
+ "epoch": 20.91,
2041
+ "learning_rate": 1.4948199309324126e-05,
2042
+ "loss": 0.0119,
2043
+ "step": 15594
2044
+ },
2045
+ {
2046
+ "epoch": 21.11,
2047
+ "learning_rate": 1.4664528860384805e-05,
2048
+ "loss": 0.009,
2049
+ "step": 15640
2050
+ },
2051
+ {
2052
+ "epoch": 21.31,
2053
+ "learning_rate": 1.4380858411445487e-05,
2054
+ "loss": 0.0113,
2055
+ "step": 15686
2056
+ },
2057
+ {
2058
+ "epoch": 21.5,
2059
+ "learning_rate": 1.4097187962506166e-05,
2060
+ "loss": 0.0096,
2061
+ "step": 15732
2062
+ },
2063
+ {
2064
+ "epoch": 21.7,
2065
+ "learning_rate": 1.3813517513566847e-05,
2066
+ "loss": 0.0103,
2067
+ "step": 15778
2068
+ },
2069
+ {
2070
+ "epoch": 21.9,
2071
+ "learning_rate": 1.3529847064627529e-05,
2072
+ "loss": 0.0089,
2073
+ "step": 15824
2074
+ },
2075
+ {
2076
+ "epoch": 22.09,
2077
+ "learning_rate": 1.324617661568821e-05,
2078
+ "loss": 0.0082,
2079
+ "step": 15870
2080
+ },
2081
+ {
2082
+ "epoch": 22.29,
2083
+ "learning_rate": 1.2962506166748891e-05,
2084
+ "loss": 0.01,
2085
+ "step": 15916
2086
+ },
2087
+ {
2088
+ "epoch": 22.49,
2089
+ "learning_rate": 1.267883571780957e-05,
2090
+ "loss": 0.0084,
2091
+ "step": 15962
2092
+ },
2093
+ {
2094
+ "epoch": 22.68,
2095
+ "learning_rate": 1.2395165268870252e-05,
2096
+ "loss": 0.0093,
2097
+ "step": 16008
2098
+ },
2099
+ {
2100
+ "epoch": 22.88,
2101
+ "learning_rate": 1.2111494819930933e-05,
2102
+ "loss": 0.0104,
2103
+ "step": 16054
2104
+ },
2105
+ {
2106
+ "epoch": 23.08,
2107
+ "learning_rate": 1.1827824370991614e-05,
2108
+ "loss": 0.0117,
2109
+ "step": 16100
2110
+ },
2111
+ {
2112
+ "epoch": 23.27,
2113
+ "learning_rate": 1.1544153922052296e-05,
2114
+ "loss": 0.0095,
2115
+ "step": 16146
2116
+ },
2117
+ {
2118
+ "epoch": 21.18,
2119
+ "learning_rate": 1.1260483473112975e-05,
2120
+ "loss": 0.0085,
2121
+ "step": 16192
2122
+ },
2123
+ {
2124
+ "epoch": 21.38,
2125
+ "learning_rate": 1.0976813024173656e-05,
2126
+ "loss": 0.0083,
2127
+ "step": 16238
2128
+ },
2129
+ {
2130
+ "epoch": 21.58,
2131
+ "learning_rate": 1.0693142575234338e-05,
2132
+ "loss": 0.0081,
2133
+ "step": 16284
2134
+ },
2135
+ {
2136
+ "epoch": 21.77,
2137
+ "learning_rate": 1.0409472126295019e-05,
2138
+ "loss": 0.0068,
2139
+ "step": 16330
2140
+ },
2141
+ {
2142
+ "epoch": 21.97,
2143
+ "learning_rate": 1.0125801677355699e-05,
2144
+ "loss": 0.0098,
2145
+ "step": 16376
2146
+ },
2147
+ {
2148
+ "epoch": 22.17,
2149
+ "learning_rate": 9.84213122841638e-06,
2150
+ "loss": 0.0087,
2151
+ "step": 16422
2152
+ },
2153
+ {
2154
+ "epoch": 22.36,
2155
+ "learning_rate": 9.558460779477061e-06,
2156
+ "loss": 0.007,
2157
+ "step": 16468
2158
+ },
2159
+ {
2160
+ "epoch": 22.56,
2161
+ "learning_rate": 9.274790330537742e-06,
2162
+ "loss": 0.0085,
2163
+ "step": 16514
2164
+ },
2165
+ {
2166
+ "epoch": 22.76,
2167
+ "learning_rate": 8.991119881598422e-06,
2168
+ "loss": 0.0094,
2169
+ "step": 16560
2170
+ },
2171
+ {
2172
+ "epoch": 22.95,
2173
+ "learning_rate": 8.707449432659103e-06,
2174
+ "loss": 0.0077,
2175
+ "step": 16606
2176
+ },
2177
+ {
2178
+ "epoch": 23.15,
2179
+ "learning_rate": 8.423778983719784e-06,
2180
+ "loss": 0.0077,
2181
+ "step": 16652
2182
+ },
2183
+ {
2184
+ "epoch": 23.35,
2185
+ "learning_rate": 8.140108534780466e-06,
2186
+ "loss": 0.008,
2187
+ "step": 16698
2188
+ },
2189
+ {
2190
+ "epoch": 23.54,
2191
+ "learning_rate": 7.856438085841145e-06,
2192
+ "loss": 0.0088,
2193
+ "step": 16744
2194
+ },
2195
+ {
2196
+ "epoch": 23.74,
2197
+ "learning_rate": 7.5727676369018255e-06,
2198
+ "loss": 0.0078,
2199
+ "step": 16790
2200
+ },
2201
+ {
2202
+ "epoch": 23.94,
2203
+ "learning_rate": 7.289097187962507e-06,
2204
+ "loss": 0.0095,
2205
+ "step": 16836
2206
+ },
2207
+ {
2208
+ "epoch": 24.13,
2209
+ "learning_rate": 7.005426739023187e-06,
2210
+ "loss": 0.0079,
2211
+ "step": 16882
2212
+ },
2213
+ {
2214
+ "epoch": 22.04,
2215
+ "learning_rate": 6.721756290083868e-06,
2216
+ "loss": 0.0068,
2217
+ "step": 16928
2218
+ },
2219
+ {
2220
+ "epoch": 22.24,
2221
+ "learning_rate": 6.438085841144549e-06,
2222
+ "loss": 0.0074,
2223
+ "step": 16974
2224
+ },
2225
+ {
2226
+ "epoch": 22.44,
2227
+ "learning_rate": 6.15441539220523e-06,
2228
+ "loss": 0.0095,
2229
+ "step": 17020
2230
+ },
2231
+ {
2232
+ "epoch": 22.63,
2233
+ "learning_rate": 5.87074494326591e-06,
2234
+ "loss": 0.007,
2235
+ "step": 17066
2236
+ },
2237
+ {
2238
+ "epoch": 22.83,
2239
+ "learning_rate": 5.587074494326592e-06,
2240
+ "loss": 0.0059,
2241
+ "step": 17112
2242
+ },
2243
+ {
2244
+ "epoch": 23.03,
2245
+ "learning_rate": 5.303404045387272e-06,
2246
+ "loss": 0.0075,
2247
+ "step": 17158
2248
+ },
2249
+ {
2250
+ "epoch": 23.22,
2251
+ "learning_rate": 5.019733596447953e-06,
2252
+ "loss": 0.0069,
2253
+ "step": 17204
2254
+ },
2255
+ {
2256
+ "epoch": 23.42,
2257
+ "learning_rate": 4.736063147508634e-06,
2258
+ "loss": 0.0068,
2259
+ "step": 17250
2260
+ },
2261
+ {
2262
+ "epoch": 23.62,
2263
+ "learning_rate": 4.452392698569315e-06,
2264
+ "loss": 0.0079,
2265
+ "step": 17296
2266
+ },
2267
+ {
2268
+ "epoch": 23.81,
2269
+ "learning_rate": 4.168722249629995e-06,
2270
+ "loss": 0.0068,
2271
+ "step": 17342
2272
+ },
2273
+ {
2274
+ "epoch": 24.01,
2275
+ "learning_rate": 3.8850518006906765e-06,
2276
+ "loss": 0.0074,
2277
+ "step": 17388
2278
+ },
2279
+ {
2280
+ "epoch": 24.21,
2281
+ "learning_rate": 3.6013813517513565e-06,
2282
+ "loss": 0.0068,
2283
+ "step": 17434
2284
+ },
2285
+ {
2286
+ "epoch": 24.4,
2287
+ "learning_rate": 3.3177109028120377e-06,
2288
+ "loss": 0.0076,
2289
+ "step": 17480
2290
+ },
2291
+ {
2292
+ "epoch": 24.6,
2293
+ "learning_rate": 3.0340404538727186e-06,
2294
+ "loss": 0.0063,
2295
+ "step": 17526
2296
+ },
2297
+ {
2298
+ "epoch": 24.79,
2299
+ "learning_rate": 2.7503700049333994e-06,
2300
+ "loss": 0.0066,
2301
+ "step": 17572
2302
+ },
2303
+ {
2304
+ "epoch": 24.99,
2305
+ "learning_rate": 2.46669955599408e-06,
2306
+ "loss": 0.0077,
2307
+ "step": 17618
2308
+ },
2309
+ {
2310
+ "epoch": 25.19,
2311
+ "learning_rate": 2.183029107054761e-06,
2312
+ "loss": 0.0077,
2313
+ "step": 17664
2314
+ },
2315
+ {
2316
+ "epoch": 23.1,
2317
+ "learning_rate": 1.8993586581154416e-06,
2318
+ "loss": 0.0065,
2319
+ "step": 17710
2320
+ },
2321
+ {
2322
+ "epoch": 23.29,
2323
+ "learning_rate": 1.6156882091761224e-06,
2324
+ "loss": 0.007,
2325
+ "step": 17756
2326
+ },
2327
+ {
2328
+ "epoch": 23.49,
2329
+ "learning_rate": 1.3320177602368033e-06,
2330
+ "loss": 0.0064,
2331
+ "step": 17802
2332
+ },
2333
+ {
2334
+ "epoch": 23.69,
2335
+ "learning_rate": 1.048347311297484e-06,
2336
+ "loss": 0.0066,
2337
+ "step": 17848
2338
+ },
2339
+ {
2340
+ "epoch": 23.88,
2341
+ "learning_rate": 7.646768623581648e-07,
2342
+ "loss": 0.0057,
2343
+ "step": 17894
2344
+ },
2345
+ {
2346
+ "epoch": 24.08,
2347
+ "learning_rate": 4.810064134188456e-07,
2348
+ "loss": 0.0063,
2349
+ "step": 17940
2350
+ },
2351
+ {
2352
+ "epoch": 24.28,
2353
+ "learning_rate": 1.973359644795264e-07,
2354
+ "loss": 0.0057,
2355
+ "step": 17986
2356
+ }
2357
+ ],
2358
+ "logging_steps": 46,
2359
+ "max_steps": 18018,
2360
+ "num_train_epochs": 77,
2361
+ "save_steps": 500,
2362
+ "total_flos": 7.874587036609413e+17,
2363
+ "trial_name": null,
2364
+ "trial_params": null
2365
+ }
checkpoint-18018/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2055871dbecc70346647edfdb5d5492b3fdb0fe7a77049f124f13d8a14abc1a5
3
+ size 4091
runs/Oct03_16-54-05_AM-3080TI-VR-PC/events.out.tfevents.1696348445.AM-3080TI-VR-PC.708.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:177d24ca530d362394c600d0d4da2fac58ca7498458ea9abfbd551a667441163
3
+ size 4184
runs/Oct03_16-56-07_AM-3080TI-VR-PC/events.out.tfevents.1696348567.AM-3080TI-VR-PC.769.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca3d4658bef345173c3becb42facc7b1fab1c975595d0c6f7b73031c0bf9c918
3
+ size 4184
runs/Oct03_16-57-18_AM-3080TI-VR-PC/events.out.tfevents.1696348638.AM-3080TI-VR-PC.827.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0317a34f7fa8938658a9f62befc65d64b995e8603565598fef25cb54bcf61d32
3
+ size 4184
runs/Oct03_16-58-27_AM-3080TI-VR-PC/events.out.tfevents.1696348708.AM-3080TI-VR-PC.884.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46e9213af42b91f49d435da52447e4dfbf33c5fe6449ea7a60bfe7501b03e443
3
+ size 66390
special_tokens_map.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<unk>",
4
+ "<s>",
5
+ "</s>"
6
+ ],
7
+ "bos_token": "<s>",
8
+ "eos_token": "</s>",
9
+ "pad_token": "</s>",
10
+ "unk_token": "<unk>"
11
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
tokenizer_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<unk>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<s>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ }
27
+ },
28
+ "additional_special_tokens": [
29
+ "<unk>",
30
+ "<s>",
31
+ "</s>"
32
+ ],
33
+ "bos_token": "<s>",
34
+ "clean_up_tokenization_spaces": false,
35
+ "eos_token": "</s>",
36
+ "legacy": true,
37
+ "model_max_length": 1024,
38
+ "pad_token": "</s>",
39
+ "sp_model_kwargs": {},
40
+ "spaces_between_special_tokens": false,
41
+ "tokenizer_class": "LlamaTokenizer",
42
+ "unk_token": "<unk>",
43
+ "use_default_system_prompt": true
44
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2055871dbecc70346647edfdb5d5492b3fdb0fe7a77049f124f13d8a14abc1a5
3
+ size 4091
training_params.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"model": "bn22/Mistral-7B-Instruct-v0.1-sharded", "data_path": "YL95/AllFpML5.10", "project_name": "mistral_naive_allFpML_v_proj", "train_split": "train", "valid_split": null, "text_column": "text", "lr": 0.0001, "epochs": 77, "batch_size": 1, "warmup_ratio": 0.1, "gradient_accumulation": 1, "optimizer": "adamw_torch", "scheduler": "linear", "weight_decay": 0.0, "max_grad_norm": 1.0, "seed": 95, "add_eos_token": false, "block_size": -1, "use_peft": true, "lora_r": 16, "lora_alpha": 32, "lora_dropout": 0.05, "logging_steps": -1, "evaluation_strategy": "epoch", "save_total_limit": 1, "save_strategy": "epoch", "auto_find_batch_size": false, "fp16": false, "push_to_hub": true, "use_int8": false, "model_max_length": 1024, "repo_id": "YL95/mistral_naive_allFpML_v_proj_3080ti", "use_int4": true, "trainer": "sft", "target_modules": "v_proj", "merge_adapter": false, "username": "YL95", "use_flash_attention_2": false}