Hugging Face's logo

HydraLM
/
mistralic-expert-31

Model card Files Community
pharaouk commited on
Commit
c5df95f
•
1 Parent(s): 87bc2c7

Training in progress, step 600, checkpoint

Browse files
Files changed (15) hide show
  1. checkpoint-600/README.md +21 -0
  2. checkpoint-600/adapter_config.json +26 -0
  3. checkpoint-600/adapter_model.bin +3 -0
  4. checkpoint-600/adapter_model/adapter_model/README.md +21 -0
  5. checkpoint-600/adapter_model/adapter_model/adapter_config.json +26 -0
  6. checkpoint-600/adapter_model/adapter_model/adapter_model.bin +3 -0
  7. checkpoint-600/added_tokens.json +8 -0
  8. checkpoint-600/optimizer.pt +3 -0
  9. checkpoint-600/rng_state.pth +3 -0
  10. checkpoint-600/scheduler.pt +3 -0
  11. checkpoint-600/special_tokens_map.json +11 -0
  12. checkpoint-600/tokenizer.model +3 -0
  13. checkpoint-600/tokenizer_config.json +74 -0
  14. checkpoint-600/trainer_state.json +3832 -0
  15. checkpoint-600/training_args.bin +3 -0
checkpoint-600/README.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: True
17
+ - bnb_4bit_compute_dtype: bfloat16
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.4.0
checkpoint-600/adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "SkunkworksAI/Mistralic-7B-1",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 16.0,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 64,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "o_proj",
18
+ "down_proj",
19
+ "q_proj",
20
+ "v_proj",
21
+ "k_proj",
22
+ "gate_proj",
23
+ "up_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
checkpoint-600/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:faf512becb2963f3aaa1bb4b7a46e182cef4dc1f6308575a69a6aab1d6386a4e
3
+ size 335706314
checkpoint-600/adapter_model/adapter_model/README.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: True
17
+ - bnb_4bit_compute_dtype: bfloat16
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.4.0
checkpoint-600/adapter_model/adapter_model/adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "SkunkworksAI/Mistralic-7B-1",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 16.0,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 64,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "o_proj",
18
+ "down_proj",
19
+ "q_proj",
20
+ "v_proj",
21
+ "k_proj",
22
+ "gate_proj",
23
+ "up_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
checkpoint-600/adapter_model/adapter_model/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:faf512becb2963f3aaa1bb4b7a46e182cef4dc1f6308575a69a6aab1d6386a4e
3
+ size 335706314
checkpoint-600/added_tokens.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</s>": 2,
3
+ "<s>": 1,
4
+ "<unk>": 0,
5
+ "<|im_end|>": 32000,
6
+ "<|im_start|>": 32001,
7
+ "[PAD]": 32002
8
+ }
checkpoint-600/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a6b7d623cdb00ad284eac885a98e18bcf6936cdcd1ba582b64d8607b72ada890
3
+ size 1342453434
checkpoint-600/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94616bbd009a027f10202b4d19e50ac13f3509cdfbbb6720f8134312ba73b0ee
3
+ size 14180
checkpoint-600/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d85fd0dcc980904bdfd3763281a80094639a2b2b63af9676380dd57c838a883
3
+ size 1064
checkpoint-600/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": "[PAD]",
10
+ "unk_token": "<unk>"
11
+ }
checkpoint-600/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
checkpoint-600/tokenizer_config.json ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": true,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": true,
16
+ "normalized": false,
17
+ "rstrip": true,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": true,
24
+ "normalized": false,
25
+ "rstrip": true,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "32000": {
30
+ "content": "<|im_end|>",
31
+ "lstrip": true,
32
+ "normalized": true,
33
+ "rstrip": true,
34
+ "single_word": false,
35
+ "special": false
36
+ },
37
+ "32001": {
38
+ "content": "<|im_start|>",
39
+ "lstrip": true,
40
+ "normalized": true,
41
+ "rstrip": true,
42
+ "single_word": false,
43
+ "special": false
44
+ },
45
+ "32002": {
46
+ "content": "[PAD]",
47
+ "lstrip": true,
48
+ "normalized": false,
49
+ "rstrip": true,
50
+ "single_word": false,
51
+ "special": true
52
+ }
53
+ },
54
+ "additional_special_tokens": [
55
+ "<unk>",
56
+ "<s>",
57
+ "</s>"
58
+ ],
59
+ "bos_token": "<s>",
60
+ "clean_up_tokenization_spaces": false,
61
+ "eos_token": "</s>",
62
+ "legacy": true,
63
+ "model_max_length": 1000000000000000019884624838656,
64
+ "pad_token": "[PAD]",
65
+ "padding_side": "right",
66
+ "sp_model_kwargs": {},
67
+ "spaces_between_special_tokens": false,
68
+ "tokenizer_class": "LlamaTokenizer",
69
+ "tokenizer_file": null,
70
+ "trust_remote_code": false,
71
+ "unk_token": "<unk>",
72
+ "use_default_system_prompt": true,
73
+ "use_fast": true
74
+ }
checkpoint-600/trainer_state.json ADDED
@@ -0,0 +1,3832 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9254454374313354,
3
+ "best_model_checkpoint": "experts/mistralic-expert-31/checkpoint-600",
4
+ "epoch": 0.18512804689910523,
5
+ "eval_steps": 200,
6
+ "global_step": 600,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 0.0002,
14
+ "loss": 0.997,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.0,
19
+ "learning_rate": 0.0002,
20
+ "loss": 1.0488,
21
+ "step": 2
22
+ },
23
+ {
24
+ "epoch": 0.0,
25
+ "learning_rate": 0.0002,
26
+ "loss": 0.9417,
27
+ "step": 3
28
+ },
29
+ {
30
+ "epoch": 0.0,
31
+ "learning_rate": 0.0002,
32
+ "loss": 1.0655,
33
+ "step": 4
34
+ },
35
+ {
36
+ "epoch": 0.0,
37
+ "learning_rate": 0.0002,
38
+ "loss": 0.9966,
39
+ "step": 5
40
+ },
41
+ {
42
+ "epoch": 0.0,
43
+ "learning_rate": 0.0002,
44
+ "loss": 1.1248,
45
+ "step": 6
46
+ },
47
+ {
48
+ "epoch": 0.0,
49
+ "learning_rate": 0.0002,
50
+ "loss": 1.0554,
51
+ "step": 7
52
+ },
53
+ {
54
+ "epoch": 0.0,
55
+ "learning_rate": 0.0002,
56
+ "loss": 1.1163,
57
+ "step": 8
58
+ },
59
+ {
60
+ "epoch": 0.0,
61
+ "learning_rate": 0.0002,
62
+ "loss": 0.9191,
63
+ "step": 9
64
+ },
65
+ {
66
+ "epoch": 0.0,
67
+ "learning_rate": 0.0002,
68
+ "loss": 1.1657,
69
+ "step": 10
70
+ },
71
+ {
72
+ "epoch": 0.0,
73
+ "learning_rate": 0.0002,
74
+ "loss": 0.9985,
75
+ "step": 11
76
+ },
77
+ {
78
+ "epoch": 0.0,
79
+ "learning_rate": 0.0002,
80
+ "loss": 1.0346,
81
+ "step": 12
82
+ },
83
+ {
84
+ "epoch": 0.0,
85
+ "learning_rate": 0.0002,
86
+ "loss": 0.8882,
87
+ "step": 13
88
+ },
89
+ {
90
+ "epoch": 0.0,
91
+ "learning_rate": 0.0002,
92
+ "loss": 1.0175,
93
+ "step": 14
94
+ },
95
+ {
96
+ "epoch": 0.0,
97
+ "learning_rate": 0.0002,
98
+ "loss": 0.877,
99
+ "step": 15
100
+ },
101
+ {
102
+ "epoch": 0.0,
103
+ "learning_rate": 0.0002,
104
+ "loss": 0.8384,
105
+ "step": 16
106
+ },
107
+ {
108
+ "epoch": 0.01,
109
+ "learning_rate": 0.0002,
110
+ "loss": 1.0233,
111
+ "step": 17
112
+ },
113
+ {
114
+ "epoch": 0.01,
115
+ "learning_rate": 0.0002,
116
+ "loss": 1.0147,
117
+ "step": 18
118
+ },
119
+ {
120
+ "epoch": 0.01,
121
+ "learning_rate": 0.0002,
122
+ "loss": 1.1521,
123
+ "step": 19
124
+ },
125
+ {
126
+ "epoch": 0.01,
127
+ "learning_rate": 0.0002,
128
+ "loss": 0.9418,
129
+ "step": 20
130
+ },
131
+ {
132
+ "epoch": 0.01,
133
+ "learning_rate": 0.0002,
134
+ "loss": 1.1608,
135
+ "step": 21
136
+ },
137
+ {
138
+ "epoch": 0.01,
139
+ "learning_rate": 0.0002,
140
+ "loss": 0.9631,
141
+ "step": 22
142
+ },
143
+ {
144
+ "epoch": 0.01,
145
+ "learning_rate": 0.0002,
146
+ "loss": 1.0209,
147
+ "step": 23
148
+ },
149
+ {
150
+ "epoch": 0.01,
151
+ "learning_rate": 0.0002,
152
+ "loss": 1.0619,
153
+ "step": 24
154
+ },
155
+ {
156
+ "epoch": 0.01,
157
+ "learning_rate": 0.0002,
158
+ "loss": 1.2694,
159
+ "step": 25
160
+ },
161
+ {
162
+ "epoch": 0.01,
163
+ "learning_rate": 0.0002,
164
+ "loss": 0.9997,
165
+ "step": 26
166
+ },
167
+ {
168
+ "epoch": 0.01,
169
+ "learning_rate": 0.0002,
170
+ "loss": 0.9234,
171
+ "step": 27
172
+ },
173
+ {
174
+ "epoch": 0.01,
175
+ "learning_rate": 0.0002,
176
+ "loss": 0.9679,
177
+ "step": 28
178
+ },
179
+ {
180
+ "epoch": 0.01,
181
+ "learning_rate": 0.0002,
182
+ "loss": 0.9442,
183
+ "step": 29
184
+ },
185
+ {
186
+ "epoch": 0.01,
187
+ "learning_rate": 0.0002,
188
+ "loss": 0.9708,
189
+ "step": 30
190
+ },
191
+ {
192
+ "epoch": 0.01,
193
+ "learning_rate": 0.0002,
194
+ "loss": 1.0319,
195
+ "step": 31
196
+ },
197
+ {
198
+ "epoch": 0.01,
199
+ "learning_rate": 0.0002,
200
+ "loss": 1.0278,
201
+ "step": 32
202
+ },
203
+ {
204
+ "epoch": 0.01,
205
+ "learning_rate": 0.0002,
206
+ "loss": 1.0284,
207
+ "step": 33
208
+ },
209
+ {
210
+ "epoch": 0.01,
211
+ "learning_rate": 0.0002,
212
+ "loss": 0.8464,
213
+ "step": 34
214
+ },
215
+ {
216
+ "epoch": 0.01,
217
+ "learning_rate": 0.0002,
218
+ "loss": 0.9889,
219
+ "step": 35
220
+ },
221
+ {
222
+ "epoch": 0.01,
223
+ "learning_rate": 0.0002,
224
+ "loss": 1.3341,
225
+ "step": 36
226
+ },
227
+ {
228
+ "epoch": 0.01,
229
+ "learning_rate": 0.0002,
230
+ "loss": 1.0814,
231
+ "step": 37
232
+ },
233
+ {
234
+ "epoch": 0.01,
235
+ "learning_rate": 0.0002,
236
+ "loss": 0.9036,
237
+ "step": 38
238
+ },
239
+ {
240
+ "epoch": 0.01,
241
+ "learning_rate": 0.0002,
242
+ "loss": 1.1541,
243
+ "step": 39
244
+ },
245
+ {
246
+ "epoch": 0.01,
247
+ "learning_rate": 0.0002,
248
+ "loss": 1.1549,
249
+ "step": 40
250
+ },
251
+ {
252
+ "epoch": 0.01,
253
+ "learning_rate": 0.0002,
254
+ "loss": 1.0072,
255
+ "step": 41
256
+ },
257
+ {
258
+ "epoch": 0.01,
259
+ "learning_rate": 0.0002,
260
+ "loss": 1.0638,
261
+ "step": 42
262
+ },
263
+ {
264
+ "epoch": 0.01,
265
+ "learning_rate": 0.0002,
266
+ "loss": 1.0451,
267
+ "step": 43
268
+ },
269
+ {
270
+ "epoch": 0.01,
271
+ "learning_rate": 0.0002,
272
+ "loss": 1.1216,
273
+ "step": 44
274
+ },
275
+ {
276
+ "epoch": 0.01,
277
+ "learning_rate": 0.0002,
278
+ "loss": 1.0063,
279
+ "step": 45
280
+ },
281
+ {
282
+ "epoch": 0.01,
283
+ "learning_rate": 0.0002,
284
+ "loss": 0.8003,
285
+ "step": 46
286
+ },
287
+ {
288
+ "epoch": 0.01,
289
+ "learning_rate": 0.0002,
290
+ "loss": 1.0479,
291
+ "step": 47
292
+ },
293
+ {
294
+ "epoch": 0.01,
295
+ "learning_rate": 0.0002,
296
+ "loss": 0.7849,
297
+ "step": 48
298
+ },
299
+ {
300
+ "epoch": 0.02,
301
+ "learning_rate": 0.0002,
302
+ "loss": 1.0547,
303
+ "step": 49
304
+ },
305
+ {
306
+ "epoch": 0.02,
307
+ "learning_rate": 0.0002,
308
+ "loss": 0.9508,
309
+ "step": 50
310
+ },
311
+ {
312
+ "epoch": 0.02,
313
+ "learning_rate": 0.0002,
314
+ "loss": 1.033,
315
+ "step": 51
316
+ },
317
+ {
318
+ "epoch": 0.02,
319
+ "learning_rate": 0.0002,
320
+ "loss": 0.7773,
321
+ "step": 52
322
+ },
323
+ {
324
+ "epoch": 0.02,
325
+ "learning_rate": 0.0002,
326
+ "loss": 0.887,
327
+ "step": 53
328
+ },
329
+ {
330
+ "epoch": 0.02,
331
+ "learning_rate": 0.0002,
332
+ "loss": 0.8354,
333
+ "step": 54
334
+ },
335
+ {
336
+ "epoch": 0.02,
337
+ "learning_rate": 0.0002,
338
+ "loss": 0.9288,
339
+ "step": 55
340
+ },
341
+ {
342
+ "epoch": 0.02,
343
+ "learning_rate": 0.0002,
344
+ "loss": 1.1684,
345
+ "step": 56
346
+ },
347
+ {
348
+ "epoch": 0.02,
349
+ "learning_rate": 0.0002,
350
+ "loss": 1.0553,
351
+ "step": 57
352
+ },
353
+ {
354
+ "epoch": 0.02,
355
+ "learning_rate": 0.0002,
356
+ "loss": 0.8693,
357
+ "step": 58
358
+ },
359
+ {
360
+ "epoch": 0.02,
361
+ "learning_rate": 0.0002,
362
+ "loss": 0.8337,
363
+ "step": 59
364
+ },
365
+ {
366
+ "epoch": 0.02,
367
+ "learning_rate": 0.0002,
368
+ "loss": 0.9965,
369
+ "step": 60
370
+ },
371
+ {
372
+ "epoch": 0.02,
373
+ "learning_rate": 0.0002,
374
+ "loss": 0.9892,
375
+ "step": 61
376
+ },
377
+ {
378
+ "epoch": 0.02,
379
+ "learning_rate": 0.0002,
380
+ "loss": 0.9308,
381
+ "step": 62
382
+ },
383
+ {
384
+ "epoch": 0.02,
385
+ "learning_rate": 0.0002,
386
+ "loss": 0.7772,
387
+ "step": 63
388
+ },
389
+ {
390
+ "epoch": 0.02,
391
+ "learning_rate": 0.0002,
392
+ "loss": 1.0411,
393
+ "step": 64
394
+ },
395
+ {
396
+ "epoch": 0.02,
397
+ "learning_rate": 0.0002,
398
+ "loss": 0.7696,
399
+ "step": 65
400
+ },
401
+ {
402
+ "epoch": 0.02,
403
+ "learning_rate": 0.0002,
404
+ "loss": 0.7961,
405
+ "step": 66
406
+ },
407
+ {
408
+ "epoch": 0.02,
409
+ "learning_rate": 0.0002,
410
+ "loss": 0.8911,
411
+ "step": 67
412
+ },
413
+ {
414
+ "epoch": 0.02,
415
+ "learning_rate": 0.0002,
416
+ "loss": 1.0532,
417
+ "step": 68
418
+ },
419
+ {
420
+ "epoch": 0.02,
421
+ "learning_rate": 0.0002,
422
+ "loss": 1.0721,
423
+ "step": 69
424
+ },
425
+ {
426
+ "epoch": 0.02,
427
+ "learning_rate": 0.0002,
428
+ "loss": 0.7499,
429
+ "step": 70
430
+ },
431
+ {
432
+ "epoch": 0.02,
433
+ "learning_rate": 0.0002,
434
+ "loss": 0.9293,
435
+ "step": 71
436
+ },
437
+ {
438
+ "epoch": 0.02,
439
+ "learning_rate": 0.0002,
440
+ "loss": 0.8478,
441
+ "step": 72
442
+ },
443
+ {
444
+ "epoch": 0.02,
445
+ "learning_rate": 0.0002,
446
+ "loss": 1.02,
447
+ "step": 73
448
+ },
449
+ {
450
+ "epoch": 0.02,
451
+ "learning_rate": 0.0002,
452
+ "loss": 1.2677,
453
+ "step": 74
454
+ },
455
+ {
456
+ "epoch": 0.02,
457
+ "learning_rate": 0.0002,
458
+ "loss": 0.9836,
459
+ "step": 75
460
+ },
461
+ {
462
+ "epoch": 0.02,
463
+ "learning_rate": 0.0002,
464
+ "loss": 1.1115,
465
+ "step": 76
466
+ },
467
+ {
468
+ "epoch": 0.02,
469
+ "learning_rate": 0.0002,
470
+ "loss": 1.1138,
471
+ "step": 77
472
+ },
473
+ {
474
+ "epoch": 0.02,
475
+ "learning_rate": 0.0002,
476
+ "loss": 0.8705,
477
+ "step": 78
478
+ },
479
+ {
480
+ "epoch": 0.02,
481
+ "learning_rate": 0.0002,
482
+ "loss": 0.9836,
483
+ "step": 79
484
+ },
485
+ {
486
+ "epoch": 0.02,
487
+ "learning_rate": 0.0002,
488
+ "loss": 1.068,
489
+ "step": 80
490
+ },
491
+ {
492
+ "epoch": 0.02,
493
+ "learning_rate": 0.0002,
494
+ "loss": 1.043,
495
+ "step": 81
496
+ },
497
+ {
498
+ "epoch": 0.03,
499
+ "learning_rate": 0.0002,
500
+ "loss": 1.0843,
501
+ "step": 82
502
+ },
503
+ {
504
+ "epoch": 0.03,
505
+ "learning_rate": 0.0002,
506
+ "loss": 0.974,
507
+ "step": 83
508
+ },
509
+ {
510
+ "epoch": 0.03,
511
+ "learning_rate": 0.0002,
512
+ "loss": 0.9724,
513
+ "step": 84
514
+ },
515
+ {
516
+ "epoch": 0.03,
517
+ "learning_rate": 0.0002,
518
+ "loss": 0.8375,
519
+ "step": 85
520
+ },
521
+ {
522
+ "epoch": 0.03,
523
+ "learning_rate": 0.0002,
524
+ "loss": 1.1525,
525
+ "step": 86
526
+ },
527
+ {
528
+ "epoch": 0.03,
529
+ "learning_rate": 0.0002,
530
+ "loss": 0.9822,
531
+ "step": 87
532
+ },
533
+ {
534
+ "epoch": 0.03,
535
+ "learning_rate": 0.0002,
536
+ "loss": 1.0685,
537
+ "step": 88
538
+ },
539
+ {
540
+ "epoch": 0.03,
541
+ "learning_rate": 0.0002,
542
+ "loss": 1.2392,
543
+ "step": 89
544
+ },
545
+ {
546
+ "epoch": 0.03,
547
+ "learning_rate": 0.0002,
548
+ "loss": 0.9015,
549
+ "step": 90
550
+ },
551
+ {
552
+ "epoch": 0.03,
553
+ "learning_rate": 0.0002,
554
+ "loss": 0.8264,
555
+ "step": 91
556
+ },
557
+ {
558
+ "epoch": 0.03,
559
+ "learning_rate": 0.0002,
560
+ "loss": 1.1202,
561
+ "step": 92
562
+ },
563
+ {
564
+ "epoch": 0.03,
565
+ "learning_rate": 0.0002,
566
+ "loss": 1.0901,
567
+ "step": 93
568
+ },
569
+ {
570
+ "epoch": 0.03,
571
+ "learning_rate": 0.0002,
572
+ "loss": 1.1147,
573
+ "step": 94
574
+ },
575
+ {
576
+ "epoch": 0.03,
577
+ "learning_rate": 0.0002,
578
+ "loss": 0.858,
579
+ "step": 95
580
+ },
581
+ {
582
+ "epoch": 0.03,
583
+ "learning_rate": 0.0002,
584
+ "loss": 1.1012,
585
+ "step": 96
586
+ },
587
+ {
588
+ "epoch": 0.03,
589
+ "learning_rate": 0.0002,
590
+ "loss": 0.9222,
591
+ "step": 97
592
+ },
593
+ {
594
+ "epoch": 0.03,
595
+ "learning_rate": 0.0002,
596
+ "loss": 1.0704,
597
+ "step": 98
598
+ },
599
+ {
600
+ "epoch": 0.03,
601
+ "learning_rate": 0.0002,
602
+ "loss": 1.1067,
603
+ "step": 99
604
+ },
605
+ {
606
+ "epoch": 0.03,
607
+ "learning_rate": 0.0002,
608
+ "loss": 0.8698,
609
+ "step": 100
610
+ },
611
+ {
612
+ "epoch": 0.03,
613
+ "learning_rate": 0.0002,
614
+ "loss": 1.2041,
615
+ "step": 101
616
+ },
617
+ {
618
+ "epoch": 0.03,
619
+ "learning_rate": 0.0002,
620
+ "loss": 0.8054,
621
+ "step": 102
622
+ },
623
+ {
624
+ "epoch": 0.03,
625
+ "learning_rate": 0.0002,
626
+ "loss": 1.0614,
627
+ "step": 103
628
+ },
629
+ {
630
+ "epoch": 0.03,
631
+ "learning_rate": 0.0002,
632
+ "loss": 0.9635,
633
+ "step": 104
634
+ },
635
+ {
636
+ "epoch": 0.03,
637
+ "learning_rate": 0.0002,
638
+ "loss": 1.0252,
639
+ "step": 105
640
+ },
641
+ {
642
+ "epoch": 0.03,
643
+ "learning_rate": 0.0002,
644
+ "loss": 0.962,
645
+ "step": 106
646
+ },
647
+ {
648
+ "epoch": 0.03,
649
+ "learning_rate": 0.0002,
650
+ "loss": 0.9954,
651
+ "step": 107
652
+ },
653
+ {
654
+ "epoch": 0.03,
655
+ "learning_rate": 0.0002,
656
+ "loss": 0.9895,
657
+ "step": 108
658
+ },
659
+ {
660
+ "epoch": 0.03,
661
+ "learning_rate": 0.0002,
662
+ "loss": 0.9363,
663
+ "step": 109
664
+ },
665
+ {
666
+ "epoch": 0.03,
667
+ "learning_rate": 0.0002,
668
+ "loss": 0.9849,
669
+ "step": 110
670
+ },
671
+ {
672
+ "epoch": 0.03,
673
+ "learning_rate": 0.0002,
674
+ "loss": 1.0117,
675
+ "step": 111
676
+ },
677
+ {
678
+ "epoch": 0.03,
679
+ "learning_rate": 0.0002,
680
+ "loss": 1.0286,
681
+ "step": 112
682
+ },
683
+ {
684
+ "epoch": 0.03,
685
+ "learning_rate": 0.0002,
686
+ "loss": 0.9685,
687
+ "step": 113
688
+ },
689
+ {
690
+ "epoch": 0.04,
691
+ "learning_rate": 0.0002,
692
+ "loss": 0.9529,
693
+ "step": 114
694
+ },
695
+ {
696
+ "epoch": 0.04,
697
+ "learning_rate": 0.0002,
698
+ "loss": 1.0492,
699
+ "step": 115
700
+ },
701
+ {
702
+ "epoch": 0.04,
703
+ "learning_rate": 0.0002,
704
+ "loss": 0.7994,
705
+ "step": 116
706
+ },
707
+ {
708
+ "epoch": 0.04,
709
+ "learning_rate": 0.0002,
710
+ "loss": 0.757,
711
+ "step": 117
712
+ },
713
+ {
714
+ "epoch": 0.04,
715
+ "learning_rate": 0.0002,
716
+ "loss": 0.9638,
717
+ "step": 118
718
+ },
719
+ {
720
+ "epoch": 0.04,
721
+ "learning_rate": 0.0002,
722
+ "loss": 1.0389,
723
+ "step": 119
724
+ },
725
+ {
726
+ "epoch": 0.04,
727
+ "learning_rate": 0.0002,
728
+ "loss": 0.9388,
729
+ "step": 120
730
+ },
731
+ {
732
+ "epoch": 0.04,
733
+ "learning_rate": 0.0002,
734
+ "loss": 0.9704,
735
+ "step": 121
736
+ },
737
+ {
738
+ "epoch": 0.04,
739
+ "learning_rate": 0.0002,
740
+ "loss": 0.8574,
741
+ "step": 122
742
+ },
743
+ {
744
+ "epoch": 0.04,
745
+ "learning_rate": 0.0002,
746
+ "loss": 1.1912,
747
+ "step": 123
748
+ },
749
+ {
750
+ "epoch": 0.04,
751
+ "learning_rate": 0.0002,
752
+ "loss": 1.0663,
753
+ "step": 124
754
+ },
755
+ {
756
+ "epoch": 0.04,
757
+ "learning_rate": 0.0002,
758
+ "loss": 1.0062,
759
+ "step": 125
760
+ },
761
+ {
762
+ "epoch": 0.04,
763
+ "learning_rate": 0.0002,
764
+ "loss": 1.0792,
765
+ "step": 126
766
+ },
767
+ {
768
+ "epoch": 0.04,
769
+ "learning_rate": 0.0002,
770
+ "loss": 0.8519,
771
+ "step": 127
772
+ },
773
+ {
774
+ "epoch": 0.04,
775
+ "learning_rate": 0.0002,
776
+ "loss": 0.913,
777
+ "step": 128
778
+ },
779
+ {
780
+ "epoch": 0.04,
781
+ "learning_rate": 0.0002,
782
+ "loss": 1.0731,
783
+ "step": 129
784
+ },
785
+ {
786
+ "epoch": 0.04,
787
+ "learning_rate": 0.0002,
788
+ "loss": 1.0924,
789
+ "step": 130
790
+ },
791
+ {
792
+ "epoch": 0.04,
793
+ "learning_rate": 0.0002,
794
+ "loss": 0.9461,
795
+ "step": 131
796
+ },
797
+ {
798
+ "epoch": 0.04,
799
+ "learning_rate": 0.0002,
800
+ "loss": 1.0235,
801
+ "step": 132
802
+ },
803
+ {
804
+ "epoch": 0.04,
805
+ "learning_rate": 0.0002,
806
+ "loss": 0.8753,
807
+ "step": 133
808
+ },
809
+ {
810
+ "epoch": 0.04,
811
+ "learning_rate": 0.0002,
812
+ "loss": 0.9494,
813
+ "step": 134
814
+ },
815
+ {
816
+ "epoch": 0.04,
817
+ "learning_rate": 0.0002,
818
+ "loss": 1.0812,
819
+ "step": 135
820
+ },
821
+ {
822
+ "epoch": 0.04,
823
+ "learning_rate": 0.0002,
824
+ "loss": 0.9319,
825
+ "step": 136
826
+ },
827
+ {
828
+ "epoch": 0.04,
829
+ "learning_rate": 0.0002,
830
+ "loss": 0.9457,
831
+ "step": 137
832
+ },
833
+ {
834
+ "epoch": 0.04,
835
+ "learning_rate": 0.0002,
836
+ "loss": 0.9424,
837
+ "step": 138
838
+ },
839
+ {
840
+ "epoch": 0.04,
841
+ "learning_rate": 0.0002,
842
+ "loss": 1.0331,
843
+ "step": 139
844
+ },
845
+ {
846
+ "epoch": 0.04,
847
+ "learning_rate": 0.0002,
848
+ "loss": 1.0472,
849
+ "step": 140
850
+ },
851
+ {
852
+ "epoch": 0.04,
853
+ "learning_rate": 0.0002,
854
+ "loss": 0.998,
855
+ "step": 141
856
+ },
857
+ {
858
+ "epoch": 0.04,
859
+ "learning_rate": 0.0002,
860
+ "loss": 0.9577,
861
+ "step": 142
862
+ },
863
+ {
864
+ "epoch": 0.04,
865
+ "learning_rate": 0.0002,
866
+ "loss": 1.1829,
867
+ "step": 143
868
+ },
869
+ {
870
+ "epoch": 0.04,
871
+ "learning_rate": 0.0002,
872
+ "loss": 1.049,
873
+ "step": 144
874
+ },
875
+ {
876
+ "epoch": 0.04,
877
+ "learning_rate": 0.0002,
878
+ "loss": 0.9152,
879
+ "step": 145
880
+ },
881
+ {
882
+ "epoch": 0.05,
883
+ "learning_rate": 0.0002,
884
+ "loss": 1.0662,
885
+ "step": 146
886
+ },
887
+ {
888
+ "epoch": 0.05,
889
+ "learning_rate": 0.0002,
890
+ "loss": 0.9938,
891
+ "step": 147
892
+ },
893
+ {
894
+ "epoch": 0.05,
895
+ "learning_rate": 0.0002,
896
+ "loss": 0.9318,
897
+ "step": 148
898
+ },
899
+ {
900
+ "epoch": 0.05,
901
+ "learning_rate": 0.0002,
902
+ "loss": 1.3438,
903
+ "step": 149
904
+ },
905
+ {
906
+ "epoch": 0.05,
907
+ "learning_rate": 0.0002,
908
+ "loss": 0.9698,
909
+ "step": 150
910
+ },
911
+ {
912
+ "epoch": 0.05,
913
+ "learning_rate": 0.0002,
914
+ "loss": 0.9604,
915
+ "step": 151
916
+ },
917
+ {
918
+ "epoch": 0.05,
919
+ "learning_rate": 0.0002,
920
+ "loss": 0.8512,
921
+ "step": 152
922
+ },
923
+ {
924
+ "epoch": 0.05,
925
+ "learning_rate": 0.0002,
926
+ "loss": 0.8416,
927
+ "step": 153
928
+ },
929
+ {
930
+ "epoch": 0.05,
931
+ "learning_rate": 0.0002,
932
+ "loss": 1.0059,
933
+ "step": 154
934
+ },
935
+ {
936
+ "epoch": 0.05,
937
+ "learning_rate": 0.0002,
938
+ "loss": 0.883,
939
+ "step": 155
940
+ },
941
+ {
942
+ "epoch": 0.05,
943
+ "learning_rate": 0.0002,
944
+ "loss": 1.2187,
945
+ "step": 156
946
+ },
947
+ {
948
+ "epoch": 0.05,
949
+ "learning_rate": 0.0002,
950
+ "loss": 0.8645,
951
+ "step": 157
952
+ },
953
+ {
954
+ "epoch": 0.05,
955
+ "learning_rate": 0.0002,
956
+ "loss": 1.0157,
957
+ "step": 158
958
+ },
959
+ {
960
+ "epoch": 0.05,
961
+ "learning_rate": 0.0002,
962
+ "loss": 1.0114,
963
+ "step": 159
964
+ },
965
+ {
966
+ "epoch": 0.05,
967
+ "learning_rate": 0.0002,
968
+ "loss": 0.9243,
969
+ "step": 160
970
+ },
971
+ {
972
+ "epoch": 0.05,
973
+ "learning_rate": 0.0002,
974
+ "loss": 0.9201,
975
+ "step": 161
976
+ },
977
+ {
978
+ "epoch": 0.05,
979
+ "learning_rate": 0.0002,
980
+ "loss": 1.1389,
981
+ "step": 162
982
+ },
983
+ {
984
+ "epoch": 0.05,
985
+ "learning_rate": 0.0002,
986
+ "loss": 0.9449,
987
+ "step": 163
988
+ },
989
+ {
990
+ "epoch": 0.05,
991
+ "learning_rate": 0.0002,
992
+ "loss": 1.0677,
993
+ "step": 164
994
+ },
995
+ {
996
+ "epoch": 0.05,
997
+ "learning_rate": 0.0002,
998
+ "loss": 1.0569,
999
+ "step": 165
1000
+ },
1001
+ {
1002
+ "epoch": 0.05,
1003
+ "learning_rate": 0.0002,
1004
+ "loss": 0.9678,
1005
+ "step": 166
1006
+ },
1007
+ {
1008
+ "epoch": 0.05,
1009
+ "learning_rate": 0.0002,
1010
+ "loss": 1.0145,
1011
+ "step": 167
1012
+ },
1013
+ {
1014
+ "epoch": 0.05,
1015
+ "learning_rate": 0.0002,
1016
+ "loss": 1.0152,
1017
+ "step": 168
1018
+ },
1019
+ {
1020
+ "epoch": 0.05,
1021
+ "learning_rate": 0.0002,
1022
+ "loss": 1.016,
1023
+ "step": 169
1024
+ },
1025
+ {
1026
+ "epoch": 0.05,
1027
+ "learning_rate": 0.0002,
1028
+ "loss": 0.8583,
1029
+ "step": 170
1030
+ },
1031
+ {
1032
+ "epoch": 0.05,
1033
+ "learning_rate": 0.0002,
1034
+ "loss": 1.073,
1035
+ "step": 171
1036
+ },
1037
+ {
1038
+ "epoch": 0.05,
1039
+ "learning_rate": 0.0002,
1040
+ "loss": 0.8966,
1041
+ "step": 172
1042
+ },
1043
+ {
1044
+ "epoch": 0.05,
1045
+ "learning_rate": 0.0002,
1046
+ "loss": 0.8771,
1047
+ "step": 173
1048
+ },
1049
+ {
1050
+ "epoch": 0.05,
1051
+ "learning_rate": 0.0002,
1052
+ "loss": 1.075,
1053
+ "step": 174
1054
+ },
1055
+ {
1056
+ "epoch": 0.05,
1057
+ "learning_rate": 0.0002,
1058
+ "loss": 0.9856,
1059
+ "step": 175
1060
+ },
1061
+ {
1062
+ "epoch": 0.05,
1063
+ "learning_rate": 0.0002,
1064
+ "loss": 0.8457,
1065
+ "step": 176
1066
+ },
1067
+ {
1068
+ "epoch": 0.05,
1069
+ "learning_rate": 0.0002,
1070
+ "loss": 1.0132,
1071
+ "step": 177
1072
+ },
1073
+ {
1074
+ "epoch": 0.05,
1075
+ "learning_rate": 0.0002,
1076
+ "loss": 0.8739,
1077
+ "step": 178
1078
+ },
1079
+ {
1080
+ "epoch": 0.06,
1081
+ "learning_rate": 0.0002,
1082
+ "loss": 0.965,
1083
+ "step": 179
1084
+ },
1085
+ {
1086
+ "epoch": 0.06,
1087
+ "learning_rate": 0.0002,
1088
+ "loss": 0.9579,
1089
+ "step": 180
1090
+ },
1091
+ {
1092
+ "epoch": 0.06,
1093
+ "learning_rate": 0.0002,
1094
+ "loss": 1.0677,
1095
+ "step": 181
1096
+ },
1097
+ {
1098
+ "epoch": 0.06,
1099
+ "learning_rate": 0.0002,
1100
+ "loss": 1.0182,
1101
+ "step": 182
1102
+ },
1103
+ {
1104
+ "epoch": 0.06,
1105
+ "learning_rate": 0.0002,
1106
+ "loss": 0.7834,
1107
+ "step": 183
1108
+ },
1109
+ {
1110
+ "epoch": 0.06,
1111
+ "learning_rate": 0.0002,
1112
+ "loss": 1.1045,
1113
+ "step": 184
1114
+ },
1115
+ {
1116
+ "epoch": 0.06,
1117
+ "learning_rate": 0.0002,
1118
+ "loss": 0.8176,
1119
+ "step": 185
1120
+ },
1121
+ {
1122
+ "epoch": 0.06,
1123
+ "learning_rate": 0.0002,
1124
+ "loss": 0.7513,
1125
+ "step": 186
1126
+ },
1127
+ {
1128
+ "epoch": 0.06,
1129
+ "learning_rate": 0.0002,
1130
+ "loss": 0.7464,
1131
+ "step": 187
1132
+ },
1133
+ {
1134
+ "epoch": 0.06,
1135
+ "learning_rate": 0.0002,
1136
+ "loss": 1.0779,
1137
+ "step": 188
1138
+ },
1139
+ {
1140
+ "epoch": 0.06,
1141
+ "learning_rate": 0.0002,
1142
+ "loss": 0.9206,
1143
+ "step": 189
1144
+ },
1145
+ {
1146
+ "epoch": 0.06,
1147
+ "learning_rate": 0.0002,
1148
+ "loss": 0.8617,
1149
+ "step": 190
1150
+ },
1151
+ {
1152
+ "epoch": 0.06,
1153
+ "learning_rate": 0.0002,
1154
+ "loss": 1.0357,
1155
+ "step": 191
1156
+ },
1157
+ {
1158
+ "epoch": 0.06,
1159
+ "learning_rate": 0.0002,
1160
+ "loss": 1.0274,
1161
+ "step": 192
1162
+ },
1163
+ {
1164
+ "epoch": 0.06,
1165
+ "learning_rate": 0.0002,
1166
+ "loss": 0.9239,
1167
+ "step": 193
1168
+ },
1169
+ {
1170
+ "epoch": 0.06,
1171
+ "learning_rate": 0.0002,
1172
+ "loss": 1.0465,
1173
+ "step": 194
1174
+ },
1175
+ {
1176
+ "epoch": 0.06,
1177
+ "learning_rate": 0.0002,
1178
+ "loss": 0.9282,
1179
+ "step": 195
1180
+ },
1181
+ {
1182
+ "epoch": 0.06,
1183
+ "learning_rate": 0.0002,
1184
+ "loss": 1.0537,
1185
+ "step": 196
1186
+ },
1187
+ {
1188
+ "epoch": 0.06,
1189
+ "learning_rate": 0.0002,
1190
+ "loss": 1.1715,
1191
+ "step": 197
1192
+ },
1193
+ {
1194
+ "epoch": 0.06,
1195
+ "learning_rate": 0.0002,
1196
+ "loss": 0.9823,
1197
+ "step": 198
1198
+ },
1199
+ {
1200
+ "epoch": 0.06,
1201
+ "learning_rate": 0.0002,
1202
+ "loss": 1.0934,
1203
+ "step": 199
1204
+ },
1205
+ {
1206
+ "epoch": 0.06,
1207
+ "learning_rate": 0.0002,
1208
+ "loss": 0.8832,
1209
+ "step": 200
1210
+ },
1211
+ {
1212
+ "epoch": 0.06,
1213
+ "eval_loss": 0.9499853253364563,
1214
+ "eval_runtime": 90.0334,
1215
+ "eval_samples_per_second": 11.107,
1216
+ "eval_steps_per_second": 5.553,
1217
+ "step": 200
1218
+ },
1219
+ {
1220
+ "epoch": 0.06,
1221
+ "mmlu_eval_accuracy": 0.583239780938983,
1222
+ "mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
1223
+ "mmlu_eval_accuracy_anatomy": 0.5,
1224
+ "mmlu_eval_accuracy_astronomy": 0.6875,
1225
+ "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
1226
+ "mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
1227
+ "mmlu_eval_accuracy_college_biology": 0.625,
1228
+ "mmlu_eval_accuracy_college_chemistry": 0.25,
1229
+ "mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
1230
+ "mmlu_eval_accuracy_college_mathematics": 0.6363636363636364,
1231
+ "mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
1232
+ "mmlu_eval_accuracy_college_physics": 0.36363636363636365,
1233
+ "mmlu_eval_accuracy_computer_security": 0.5454545454545454,
1234
+ "mmlu_eval_accuracy_conceptual_physics": 0.5384615384615384,
1235
+ "mmlu_eval_accuracy_econometrics": 0.5,
1236
+ "mmlu_eval_accuracy_electrical_engineering": 0.625,
1237
+ "mmlu_eval_accuracy_elementary_mathematics": 0.43902439024390244,
1238
+ "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
1239
+ "mmlu_eval_accuracy_global_facts": 0.4,
1240
+ "mmlu_eval_accuracy_high_school_biology": 0.53125,
1241
+ "mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
1242
+ "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
1243
+ "mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
1244
+ "mmlu_eval_accuracy_high_school_geography": 0.7727272727272727,
1245
+ "mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
1246
+ "mmlu_eval_accuracy_high_school_macroeconomics": 0.5813953488372093,
1247
+ "mmlu_eval_accuracy_high_school_mathematics": 0.3448275862068966,
1248
+ "mmlu_eval_accuracy_high_school_microeconomics": 0.5769230769230769,
1249
+ "mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
1250
+ "mmlu_eval_accuracy_high_school_psychology": 0.85,
1251
+ "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
1252
+ "mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
1253
+ "mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
1254
+ "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
1255
+ "mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
1256
+ "mmlu_eval_accuracy_international_law": 0.9230769230769231,
1257
+ "mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
1258
+ "mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
1259
+ "mmlu_eval_accuracy_machine_learning": 0.45454545454545453,
1260
+ "mmlu_eval_accuracy_management": 0.9090909090909091,
1261
+ "mmlu_eval_accuracy_marketing": 0.92,
1262
+ "mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
1263
+ "mmlu_eval_accuracy_miscellaneous": 0.7674418604651163,
1264
+ "mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
1265
+ "mmlu_eval_accuracy_moral_scenarios": 0.28,
1266
+ "mmlu_eval_accuracy_nutrition": 0.6666666666666666,
1267
+ "mmlu_eval_accuracy_philosophy": 0.7647058823529411,
1268
+ "mmlu_eval_accuracy_prehistory": 0.45714285714285713,
1269
+ "mmlu_eval_accuracy_professional_accounting": 0.5806451612903226,
1270
+ "mmlu_eval_accuracy_professional_law": 0.3941176470588235,
1271
+ "mmlu_eval_accuracy_professional_medicine": 0.6129032258064516,
1272
+ "mmlu_eval_accuracy_professional_psychology": 0.6521739130434783,
1273
+ "mmlu_eval_accuracy_public_relations": 0.5,
1274
+ "mmlu_eval_accuracy_security_studies": 0.6666666666666666,
1275
+ "mmlu_eval_accuracy_sociology": 0.8636363636363636,
1276
+ "mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
1277
+ "mmlu_eval_accuracy_virology": 0.5555555555555556,
1278
+ "mmlu_eval_accuracy_world_religions": 0.8421052631578947,
1279
+ "mmlu_loss": 1.5736223915227088,
1280
+ "step": 200
1281
+ },
1282
+ {
1283
+ "epoch": 0.06,
1284
+ "learning_rate": 0.0002,
1285
+ "loss": 0.925,
1286
+ "step": 201
1287
+ },
1288
+ {
1289
+ "epoch": 0.06,
1290
+ "learning_rate": 0.0002,
1291
+ "loss": 0.9671,
1292
+ "step": 202
1293
+ },
1294
+ {
1295
+ "epoch": 0.06,
1296
+ "learning_rate": 0.0002,
1297
+ "loss": 0.8345,
1298
+ "step": 203
1299
+ },
1300
+ {
1301
+ "epoch": 0.06,
1302
+ "learning_rate": 0.0002,
1303
+ "loss": 0.8449,
1304
+ "step": 204
1305
+ },
1306
+ {
1307
+ "epoch": 0.06,
1308
+ "learning_rate": 0.0002,
1309
+ "loss": 0.9823,
1310
+ "step": 205
1311
+ },
1312
+ {
1313
+ "epoch": 0.06,
1314
+ "learning_rate": 0.0002,
1315
+ "loss": 1.1645,
1316
+ "step": 206
1317
+ },
1318
+ {
1319
+ "epoch": 0.06,
1320
+ "learning_rate": 0.0002,
1321
+ "loss": 1.0478,
1322
+ "step": 207
1323
+ },
1324
+ {
1325
+ "epoch": 0.06,
1326
+ "learning_rate": 0.0002,
1327
+ "loss": 1.0589,
1328
+ "step": 208
1329
+ },
1330
+ {
1331
+ "epoch": 0.06,
1332
+ "learning_rate": 0.0002,
1333
+ "loss": 0.9737,
1334
+ "step": 209
1335
+ },
1336
+ {
1337
+ "epoch": 0.06,
1338
+ "learning_rate": 0.0002,
1339
+ "loss": 1.0603,
1340
+ "step": 210
1341
+ },
1342
+ {
1343
+ "epoch": 0.07,
1344
+ "learning_rate": 0.0002,
1345
+ "loss": 1.0362,
1346
+ "step": 211
1347
+ },
1348
+ {
1349
+ "epoch": 0.07,
1350
+ "learning_rate": 0.0002,
1351
+ "loss": 0.9799,
1352
+ "step": 212
1353
+ },
1354
+ {
1355
+ "epoch": 0.07,
1356
+ "learning_rate": 0.0002,
1357
+ "loss": 0.8978,
1358
+ "step": 213
1359
+ },
1360
+ {
1361
+ "epoch": 0.07,
1362
+ "learning_rate": 0.0002,
1363
+ "loss": 0.9695,
1364
+ "step": 214
1365
+ },
1366
+ {
1367
+ "epoch": 0.07,
1368
+ "learning_rate": 0.0002,
1369
+ "loss": 0.9839,
1370
+ "step": 215
1371
+ },
1372
+ {
1373
+ "epoch": 0.07,
1374
+ "learning_rate": 0.0002,
1375
+ "loss": 1.0147,
1376
+ "step": 216
1377
+ },
1378
+ {
1379
+ "epoch": 0.07,
1380
+ "learning_rate": 0.0002,
1381
+ "loss": 0.9018,
1382
+ "step": 217
1383
+ },
1384
+ {
1385
+ "epoch": 0.07,
1386
+ "learning_rate": 0.0002,
1387
+ "loss": 1.1482,
1388
+ "step": 218
1389
+ },
1390
+ {
1391
+ "epoch": 0.07,
1392
+ "learning_rate": 0.0002,
1393
+ "loss": 0.9104,
1394
+ "step": 219
1395
+ },
1396
+ {
1397
+ "epoch": 0.07,
1398
+ "learning_rate": 0.0002,
1399
+ "loss": 1.0003,
1400
+ "step": 220
1401
+ },
1402
+ {
1403
+ "epoch": 0.07,
1404
+ "learning_rate": 0.0002,
1405
+ "loss": 0.8485,
1406
+ "step": 221
1407
+ },
1408
+ {
1409
+ "epoch": 0.07,
1410
+ "learning_rate": 0.0002,
1411
+ "loss": 0.8124,
1412
+ "step": 222
1413
+ },
1414
+ {
1415
+ "epoch": 0.07,
1416
+ "learning_rate": 0.0002,
1417
+ "loss": 0.9756,
1418
+ "step": 223
1419
+ },
1420
+ {
1421
+ "epoch": 0.07,
1422
+ "learning_rate": 0.0002,
1423
+ "loss": 1.2614,
1424
+ "step": 224
1425
+ },
1426
+ {
1427
+ "epoch": 0.07,
1428
+ "learning_rate": 0.0002,
1429
+ "loss": 1.1372,
1430
+ "step": 225
1431
+ },
1432
+ {
1433
+ "epoch": 0.07,
1434
+ "learning_rate": 0.0002,
1435
+ "loss": 1.0422,
1436
+ "step": 226
1437
+ },
1438
+ {
1439
+ "epoch": 0.07,
1440
+ "learning_rate": 0.0002,
1441
+ "loss": 1.0229,
1442
+ "step": 227
1443
+ },
1444
+ {
1445
+ "epoch": 0.07,
1446
+ "learning_rate": 0.0002,
1447
+ "loss": 0.9819,
1448
+ "step": 228
1449
+ },
1450
+ {
1451
+ "epoch": 0.07,
1452
+ "learning_rate": 0.0002,
1453
+ "loss": 0.917,
1454
+ "step": 229
1455
+ },
1456
+ {
1457
+ "epoch": 0.07,
1458
+ "learning_rate": 0.0002,
1459
+ "loss": 1.0221,
1460
+ "step": 230
1461
+ },
1462
+ {
1463
+ "epoch": 0.07,
1464
+ "learning_rate": 0.0002,
1465
+ "loss": 1.0757,
1466
+ "step": 231
1467
+ },
1468
+ {
1469
+ "epoch": 0.07,
1470
+ "learning_rate": 0.0002,
1471
+ "loss": 1.1495,
1472
+ "step": 232
1473
+ },
1474
+ {
1475
+ "epoch": 0.07,
1476
+ "learning_rate": 0.0002,
1477
+ "loss": 1.019,
1478
+ "step": 233
1479
+ },
1480
+ {
1481
+ "epoch": 0.07,
1482
+ "learning_rate": 0.0002,
1483
+ "loss": 0.9665,
1484
+ "step": 234
1485
+ },
1486
+ {
1487
+ "epoch": 0.07,
1488
+ "learning_rate": 0.0002,
1489
+ "loss": 1.0675,
1490
+ "step": 235
1491
+ },
1492
+ {
1493
+ "epoch": 0.07,
1494
+ "learning_rate": 0.0002,
1495
+ "loss": 0.8124,
1496
+ "step": 236
1497
+ },
1498
+ {
1499
+ "epoch": 0.07,
1500
+ "learning_rate": 0.0002,
1501
+ "loss": 1.0367,
1502
+ "step": 237
1503
+ },
1504
+ {
1505
+ "epoch": 0.07,
1506
+ "learning_rate": 0.0002,
1507
+ "loss": 1.001,
1508
+ "step": 238
1509
+ },
1510
+ {
1511
+ "epoch": 0.07,
1512
+ "learning_rate": 0.0002,
1513
+ "loss": 0.7586,
1514
+ "step": 239
1515
+ },
1516
+ {
1517
+ "epoch": 0.07,
1518
+ "learning_rate": 0.0002,
1519
+ "loss": 1.0411,
1520
+ "step": 240
1521
+ },
1522
+ {
1523
+ "epoch": 0.07,
1524
+ "learning_rate": 0.0002,
1525
+ "loss": 0.8661,
1526
+ "step": 241
1527
+ },
1528
+ {
1529
+ "epoch": 0.07,
1530
+ "learning_rate": 0.0002,
1531
+ "loss": 0.9123,
1532
+ "step": 242
1533
+ },
1534
+ {
1535
+ "epoch": 0.07,
1536
+ "learning_rate": 0.0002,
1537
+ "loss": 0.83,
1538
+ "step": 243
1539
+ },
1540
+ {
1541
+ "epoch": 0.08,
1542
+ "learning_rate": 0.0002,
1543
+ "loss": 0.8899,
1544
+ "step": 244
1545
+ },
1546
+ {
1547
+ "epoch": 0.08,
1548
+ "learning_rate": 0.0002,
1549
+ "loss": 1.0077,
1550
+ "step": 245
1551
+ },
1552
+ {
1553
+ "epoch": 0.08,
1554
+ "learning_rate": 0.0002,
1555
+ "loss": 0.9373,
1556
+ "step": 246
1557
+ },
1558
+ {
1559
+ "epoch": 0.08,
1560
+ "learning_rate": 0.0002,
1561
+ "loss": 1.1275,
1562
+ "step": 247
1563
+ },
1564
+ {
1565
+ "epoch": 0.08,
1566
+ "learning_rate": 0.0002,
1567
+ "loss": 0.7653,
1568
+ "step": 248
1569
+ },
1570
+ {
1571
+ "epoch": 0.08,
1572
+ "learning_rate": 0.0002,
1573
+ "loss": 0.8757,
1574
+ "step": 249
1575
+ },
1576
+ {
1577
+ "epoch": 0.08,
1578
+ "learning_rate": 0.0002,
1579
+ "loss": 1.1311,
1580
+ "step": 250
1581
+ },
1582
+ {
1583
+ "epoch": 0.08,
1584
+ "learning_rate": 0.0002,
1585
+ "loss": 0.8657,
1586
+ "step": 251
1587
+ },
1588
+ {
1589
+ "epoch": 0.08,
1590
+ "learning_rate": 0.0002,
1591
+ "loss": 1.1071,
1592
+ "step": 252
1593
+ },
1594
+ {
1595
+ "epoch": 0.08,
1596
+ "learning_rate": 0.0002,
1597
+ "loss": 0.9579,
1598
+ "step": 253
1599
+ },
1600
+ {
1601
+ "epoch": 0.08,
1602
+ "learning_rate": 0.0002,
1603
+ "loss": 0.8067,
1604
+ "step": 254
1605
+ },
1606
+ {
1607
+ "epoch": 0.08,
1608
+ "learning_rate": 0.0002,
1609
+ "loss": 1.1368,
1610
+ "step": 255
1611
+ },
1612
+ {
1613
+ "epoch": 0.08,
1614
+ "learning_rate": 0.0002,
1615
+ "loss": 0.8264,
1616
+ "step": 256
1617
+ },
1618
+ {
1619
+ "epoch": 0.08,
1620
+ "learning_rate": 0.0002,
1621
+ "loss": 0.9937,
1622
+ "step": 257
1623
+ },
1624
+ {
1625
+ "epoch": 0.08,
1626
+ "learning_rate": 0.0002,
1627
+ "loss": 1.0458,
1628
+ "step": 258
1629
+ },
1630
+ {
1631
+ "epoch": 0.08,
1632
+ "learning_rate": 0.0002,
1633
+ "loss": 0.9694,
1634
+ "step": 259
1635
+ },
1636
+ {
1637
+ "epoch": 0.08,
1638
+ "learning_rate": 0.0002,
1639
+ "loss": 1.0525,
1640
+ "step": 260
1641
+ },
1642
+ {
1643
+ "epoch": 0.08,
1644
+ "learning_rate": 0.0002,
1645
+ "loss": 0.8832,
1646
+ "step": 261
1647
+ },
1648
+ {
1649
+ "epoch": 0.08,
1650
+ "learning_rate": 0.0002,
1651
+ "loss": 0.8712,
1652
+ "step": 262
1653
+ },
1654
+ {
1655
+ "epoch": 0.08,
1656
+ "learning_rate": 0.0002,
1657
+ "loss": 1.0039,
1658
+ "step": 263
1659
+ },
1660
+ {
1661
+ "epoch": 0.08,
1662
+ "learning_rate": 0.0002,
1663
+ "loss": 1.1049,
1664
+ "step": 264
1665
+ },
1666
+ {
1667
+ "epoch": 0.08,
1668
+ "learning_rate": 0.0002,
1669
+ "loss": 0.7728,
1670
+ "step": 265
1671
+ },
1672
+ {
1673
+ "epoch": 0.08,
1674
+ "learning_rate": 0.0002,
1675
+ "loss": 0.8612,
1676
+ "step": 266
1677
+ },
1678
+ {
1679
+ "epoch": 0.08,
1680
+ "learning_rate": 0.0002,
1681
+ "loss": 0.9169,
1682
+ "step": 267
1683
+ },
1684
+ {
1685
+ "epoch": 0.08,
1686
+ "learning_rate": 0.0002,
1687
+ "loss": 0.9523,
1688
+ "step": 268
1689
+ },
1690
+ {
1691
+ "epoch": 0.08,
1692
+ "learning_rate": 0.0002,
1693
+ "loss": 0.8041,
1694
+ "step": 269
1695
+ },
1696
+ {
1697
+ "epoch": 0.08,
1698
+ "learning_rate": 0.0002,
1699
+ "loss": 0.8771,
1700
+ "step": 270
1701
+ },
1702
+ {
1703
+ "epoch": 0.08,
1704
+ "learning_rate": 0.0002,
1705
+ "loss": 1.1153,
1706
+ "step": 271
1707
+ },
1708
+ {
1709
+ "epoch": 0.08,
1710
+ "learning_rate": 0.0002,
1711
+ "loss": 0.8168,
1712
+ "step": 272
1713
+ },
1714
+ {
1715
+ "epoch": 0.08,
1716
+ "learning_rate": 0.0002,
1717
+ "loss": 0.796,
1718
+ "step": 273
1719
+ },
1720
+ {
1721
+ "epoch": 0.08,
1722
+ "learning_rate": 0.0002,
1723
+ "loss": 0.9457,
1724
+ "step": 274
1725
+ },
1726
+ {
1727
+ "epoch": 0.08,
1728
+ "learning_rate": 0.0002,
1729
+ "loss": 0.9758,
1730
+ "step": 275
1731
+ },
1732
+ {
1733
+ "epoch": 0.09,
1734
+ "learning_rate": 0.0002,
1735
+ "loss": 1.1362,
1736
+ "step": 276
1737
+ },
1738
+ {
1739
+ "epoch": 0.09,
1740
+ "learning_rate": 0.0002,
1741
+ "loss": 0.8275,
1742
+ "step": 277
1743
+ },
1744
+ {
1745
+ "epoch": 0.09,
1746
+ "learning_rate": 0.0002,
1747
+ "loss": 0.8748,
1748
+ "step": 278
1749
+ },
1750
+ {
1751
+ "epoch": 0.09,
1752
+ "learning_rate": 0.0002,
1753
+ "loss": 0.9961,
1754
+ "step": 279
1755
+ },
1756
+ {
1757
+ "epoch": 0.09,
1758
+ "learning_rate": 0.0002,
1759
+ "loss": 0.8587,
1760
+ "step": 280
1761
+ },
1762
+ {
1763
+ "epoch": 0.09,
1764
+ "learning_rate": 0.0002,
1765
+ "loss": 0.9617,
1766
+ "step": 281
1767
+ },
1768
+ {
1769
+ "epoch": 0.09,
1770
+ "learning_rate": 0.0002,
1771
+ "loss": 1.3358,
1772
+ "step": 282
1773
+ },
1774
+ {
1775
+ "epoch": 0.09,
1776
+ "learning_rate": 0.0002,
1777
+ "loss": 0.959,
1778
+ "step": 283
1779
+ },
1780
+ {
1781
+ "epoch": 0.09,
1782
+ "learning_rate": 0.0002,
1783
+ "loss": 0.9922,
1784
+ "step": 284
1785
+ },
1786
+ {
1787
+ "epoch": 0.09,
1788
+ "learning_rate": 0.0002,
1789
+ "loss": 1.0314,
1790
+ "step": 285
1791
+ },
1792
+ {
1793
+ "epoch": 0.09,
1794
+ "learning_rate": 0.0002,
1795
+ "loss": 0.9845,
1796
+ "step": 286
1797
+ },
1798
+ {
1799
+ "epoch": 0.09,
1800
+ "learning_rate": 0.0002,
1801
+ "loss": 0.887,
1802
+ "step": 287
1803
+ },
1804
+ {
1805
+ "epoch": 0.09,
1806
+ "learning_rate": 0.0002,
1807
+ "loss": 1.0033,
1808
+ "step": 288
1809
+ },
1810
+ {
1811
+ "epoch": 0.09,
1812
+ "learning_rate": 0.0002,
1813
+ "loss": 1.0781,
1814
+ "step": 289
1815
+ },
1816
+ {
1817
+ "epoch": 0.09,
1818
+ "learning_rate": 0.0002,
1819
+ "loss": 1.1126,
1820
+ "step": 290
1821
+ },
1822
+ {
1823
+ "epoch": 0.09,
1824
+ "learning_rate": 0.0002,
1825
+ "loss": 0.9803,
1826
+ "step": 291
1827
+ },
1828
+ {
1829
+ "epoch": 0.09,
1830
+ "learning_rate": 0.0002,
1831
+ "loss": 1.0852,
1832
+ "step": 292
1833
+ },
1834
+ {
1835
+ "epoch": 0.09,
1836
+ "learning_rate": 0.0002,
1837
+ "loss": 0.7943,
1838
+ "step": 293
1839
+ },
1840
+ {
1841
+ "epoch": 0.09,
1842
+ "learning_rate": 0.0002,
1843
+ "loss": 0.9783,
1844
+ "step": 294
1845
+ },
1846
+ {
1847
+ "epoch": 0.09,
1848
+ "learning_rate": 0.0002,
1849
+ "loss": 0.9389,
1850
+ "step": 295
1851
+ },
1852
+ {
1853
+ "epoch": 0.09,
1854
+ "learning_rate": 0.0002,
1855
+ "loss": 1.0074,
1856
+ "step": 296
1857
+ },
1858
+ {
1859
+ "epoch": 0.09,
1860
+ "learning_rate": 0.0002,
1861
+ "loss": 0.7229,
1862
+ "step": 297
1863
+ },
1864
+ {
1865
+ "epoch": 0.09,
1866
+ "learning_rate": 0.0002,
1867
+ "loss": 1.0911,
1868
+ "step": 298
1869
+ },
1870
+ {
1871
+ "epoch": 0.09,
1872
+ "learning_rate": 0.0002,
1873
+ "loss": 0.952,
1874
+ "step": 299
1875
+ },
1876
+ {
1877
+ "epoch": 0.09,
1878
+ "learning_rate": 0.0002,
1879
+ "loss": 1.0237,
1880
+ "step": 300
1881
+ },
1882
+ {
1883
+ "epoch": 0.09,
1884
+ "learning_rate": 0.0002,
1885
+ "loss": 1.14,
1886
+ "step": 301
1887
+ },
1888
+ {
1889
+ "epoch": 0.09,
1890
+ "learning_rate": 0.0002,
1891
+ "loss": 0.7661,
1892
+ "step": 302
1893
+ },
1894
+ {
1895
+ "epoch": 0.09,
1896
+ "learning_rate": 0.0002,
1897
+ "loss": 0.9887,
1898
+ "step": 303
1899
+ },
1900
+ {
1901
+ "epoch": 0.09,
1902
+ "learning_rate": 0.0002,
1903
+ "loss": 1.1378,
1904
+ "step": 304
1905
+ },
1906
+ {
1907
+ "epoch": 0.09,
1908
+ "learning_rate": 0.0002,
1909
+ "loss": 1.0622,
1910
+ "step": 305
1911
+ },
1912
+ {
1913
+ "epoch": 0.09,
1914
+ "learning_rate": 0.0002,
1915
+ "loss": 0.8306,
1916
+ "step": 306
1917
+ },
1918
+ {
1919
+ "epoch": 0.09,
1920
+ "learning_rate": 0.0002,
1921
+ "loss": 0.71,
1922
+ "step": 307
1923
+ },
1924
+ {
1925
+ "epoch": 0.1,
1926
+ "learning_rate": 0.0002,
1927
+ "loss": 0.7812,
1928
+ "step": 308
1929
+ },
1930
+ {
1931
+ "epoch": 0.1,
1932
+ "learning_rate": 0.0002,
1933
+ "loss": 1.0251,
1934
+ "step": 309
1935
+ },
1936
+ {
1937
+ "epoch": 0.1,
1938
+ "learning_rate": 0.0002,
1939
+ "loss": 1.1483,
1940
+ "step": 310
1941
+ },
1942
+ {
1943
+ "epoch": 0.1,
1944
+ "learning_rate": 0.0002,
1945
+ "loss": 1.041,
1946
+ "step": 311
1947
+ },
1948
+ {
1949
+ "epoch": 0.1,
1950
+ "learning_rate": 0.0002,
1951
+ "loss": 1.0637,
1952
+ "step": 312
1953
+ },
1954
+ {
1955
+ "epoch": 0.1,
1956
+ "learning_rate": 0.0002,
1957
+ "loss": 1.0051,
1958
+ "step": 313
1959
+ },
1960
+ {
1961
+ "epoch": 0.1,
1962
+ "learning_rate": 0.0002,
1963
+ "loss": 1.021,
1964
+ "step": 314
1965
+ },
1966
+ {
1967
+ "epoch": 0.1,
1968
+ "learning_rate": 0.0002,
1969
+ "loss": 1.03,
1970
+ "step": 315
1971
+ },
1972
+ {
1973
+ "epoch": 0.1,
1974
+ "learning_rate": 0.0002,
1975
+ "loss": 1.2441,
1976
+ "step": 316
1977
+ },
1978
+ {
1979
+ "epoch": 0.1,
1980
+ "learning_rate": 0.0002,
1981
+ "loss": 0.8661,
1982
+ "step": 317
1983
+ },
1984
+ {
1985
+ "epoch": 0.1,
1986
+ "learning_rate": 0.0002,
1987
+ "loss": 0.9945,
1988
+ "step": 318
1989
+ },
1990
+ {
1991
+ "epoch": 0.1,
1992
+ "learning_rate": 0.0002,
1993
+ "loss": 1.1312,
1994
+ "step": 319
1995
+ },
1996
+ {
1997
+ "epoch": 0.1,
1998
+ "learning_rate": 0.0002,
1999
+ "loss": 1.083,
2000
+ "step": 320
2001
+ },
2002
+ {
2003
+ "epoch": 0.1,
2004
+ "learning_rate": 0.0002,
2005
+ "loss": 0.9382,
2006
+ "step": 321
2007
+ },
2008
+ {
2009
+ "epoch": 0.1,
2010
+ "learning_rate": 0.0002,
2011
+ "loss": 0.9763,
2012
+ "step": 322
2013
+ },
2014
+ {
2015
+ "epoch": 0.1,
2016
+ "learning_rate": 0.0002,
2017
+ "loss": 0.9111,
2018
+ "step": 323
2019
+ },
2020
+ {
2021
+ "epoch": 0.1,
2022
+ "learning_rate": 0.0002,
2023
+ "loss": 1.0146,
2024
+ "step": 324
2025
+ },
2026
+ {
2027
+ "epoch": 0.1,
2028
+ "learning_rate": 0.0002,
2029
+ "loss": 1.1616,
2030
+ "step": 325
2031
+ },
2032
+ {
2033
+ "epoch": 0.1,
2034
+ "learning_rate": 0.0002,
2035
+ "loss": 0.8338,
2036
+ "step": 326
2037
+ },
2038
+ {
2039
+ "epoch": 0.1,
2040
+ "learning_rate": 0.0002,
2041
+ "loss": 0.9968,
2042
+ "step": 327
2043
+ },
2044
+ {
2045
+ "epoch": 0.1,
2046
+ "learning_rate": 0.0002,
2047
+ "loss": 1.022,
2048
+ "step": 328
2049
+ },
2050
+ {
2051
+ "epoch": 0.1,
2052
+ "learning_rate": 0.0002,
2053
+ "loss": 1.0467,
2054
+ "step": 329
2055
+ },
2056
+ {
2057
+ "epoch": 0.1,
2058
+ "learning_rate": 0.0002,
2059
+ "loss": 1.188,
2060
+ "step": 330
2061
+ },
2062
+ {
2063
+ "epoch": 0.1,
2064
+ "learning_rate": 0.0002,
2065
+ "loss": 1.028,
2066
+ "step": 331
2067
+ },
2068
+ {
2069
+ "epoch": 0.1,
2070
+ "learning_rate": 0.0002,
2071
+ "loss": 0.982,
2072
+ "step": 332
2073
+ },
2074
+ {
2075
+ "epoch": 0.1,
2076
+ "learning_rate": 0.0002,
2077
+ "loss": 1.0669,
2078
+ "step": 333
2079
+ },
2080
+ {
2081
+ "epoch": 0.1,
2082
+ "learning_rate": 0.0002,
2083
+ "loss": 0.875,
2084
+ "step": 334
2085
+ },
2086
+ {
2087
+ "epoch": 0.1,
2088
+ "learning_rate": 0.0002,
2089
+ "loss": 0.9985,
2090
+ "step": 335
2091
+ },
2092
+ {
2093
+ "epoch": 0.1,
2094
+ "learning_rate": 0.0002,
2095
+ "loss": 0.9164,
2096
+ "step": 336
2097
+ },
2098
+ {
2099
+ "epoch": 0.1,
2100
+ "learning_rate": 0.0002,
2101
+ "loss": 1.0613,
2102
+ "step": 337
2103
+ },
2104
+ {
2105
+ "epoch": 0.1,
2106
+ "learning_rate": 0.0002,
2107
+ "loss": 0.9191,
2108
+ "step": 338
2109
+ },
2110
+ {
2111
+ "epoch": 0.1,
2112
+ "learning_rate": 0.0002,
2113
+ "loss": 0.8803,
2114
+ "step": 339
2115
+ },
2116
+ {
2117
+ "epoch": 0.1,
2118
+ "learning_rate": 0.0002,
2119
+ "loss": 0.8468,
2120
+ "step": 340
2121
+ },
2122
+ {
2123
+ "epoch": 0.11,
2124
+ "learning_rate": 0.0002,
2125
+ "loss": 1.0023,
2126
+ "step": 341
2127
+ },
2128
+ {
2129
+ "epoch": 0.11,
2130
+ "learning_rate": 0.0002,
2131
+ "loss": 1.0508,
2132
+ "step": 342
2133
+ },
2134
+ {
2135
+ "epoch": 0.11,
2136
+ "learning_rate": 0.0002,
2137
+ "loss": 0.83,
2138
+ "step": 343
2139
+ },
2140
+ {
2141
+ "epoch": 0.11,
2142
+ "learning_rate": 0.0002,
2143
+ "loss": 1.0438,
2144
+ "step": 344
2145
+ },
2146
+ {
2147
+ "epoch": 0.11,
2148
+ "learning_rate": 0.0002,
2149
+ "loss": 1.0424,
2150
+ "step": 345
2151
+ },
2152
+ {
2153
+ "epoch": 0.11,
2154
+ "learning_rate": 0.0002,
2155
+ "loss": 1.0642,
2156
+ "step": 346
2157
+ },
2158
+ {
2159
+ "epoch": 0.11,
2160
+ "learning_rate": 0.0002,
2161
+ "loss": 0.9585,
2162
+ "step": 347
2163
+ },
2164
+ {
2165
+ "epoch": 0.11,
2166
+ "learning_rate": 0.0002,
2167
+ "loss": 0.8884,
2168
+ "step": 348
2169
+ },
2170
+ {
2171
+ "epoch": 0.11,
2172
+ "learning_rate": 0.0002,
2173
+ "loss": 0.9328,
2174
+ "step": 349
2175
+ },
2176
+ {
2177
+ "epoch": 0.11,
2178
+ "learning_rate": 0.0002,
2179
+ "loss": 0.8091,
2180
+ "step": 350
2181
+ },
2182
+ {
2183
+ "epoch": 0.11,
2184
+ "learning_rate": 0.0002,
2185
+ "loss": 1.0103,
2186
+ "step": 351
2187
+ },
2188
+ {
2189
+ "epoch": 0.11,
2190
+ "learning_rate": 0.0002,
2191
+ "loss": 1.2493,
2192
+ "step": 352
2193
+ },
2194
+ {
2195
+ "epoch": 0.11,
2196
+ "learning_rate": 0.0002,
2197
+ "loss": 0.9404,
2198
+ "step": 353
2199
+ },
2200
+ {
2201
+ "epoch": 0.11,
2202
+ "learning_rate": 0.0002,
2203
+ "loss": 0.9181,
2204
+ "step": 354
2205
+ },
2206
+ {
2207
+ "epoch": 0.11,
2208
+ "learning_rate": 0.0002,
2209
+ "loss": 0.8925,
2210
+ "step": 355
2211
+ },
2212
+ {
2213
+ "epoch": 0.11,
2214
+ "learning_rate": 0.0002,
2215
+ "loss": 0.8298,
2216
+ "step": 356
2217
+ },
2218
+ {
2219
+ "epoch": 0.11,
2220
+ "learning_rate": 0.0002,
2221
+ "loss": 0.9405,
2222
+ "step": 357
2223
+ },
2224
+ {
2225
+ "epoch": 0.11,
2226
+ "learning_rate": 0.0002,
2227
+ "loss": 0.9222,
2228
+ "step": 358
2229
+ },
2230
+ {
2231
+ "epoch": 0.11,
2232
+ "learning_rate": 0.0002,
2233
+ "loss": 1.0356,
2234
+ "step": 359
2235
+ },
2236
+ {
2237
+ "epoch": 0.11,
2238
+ "learning_rate": 0.0002,
2239
+ "loss": 1.0183,
2240
+ "step": 360
2241
+ },
2242
+ {
2243
+ "epoch": 0.11,
2244
+ "learning_rate": 0.0002,
2245
+ "loss": 0.8807,
2246
+ "step": 361
2247
+ },
2248
+ {
2249
+ "epoch": 0.11,
2250
+ "learning_rate": 0.0002,
2251
+ "loss": 0.9507,
2252
+ "step": 362
2253
+ },
2254
+ {
2255
+ "epoch": 0.11,
2256
+ "learning_rate": 0.0002,
2257
+ "loss": 0.9158,
2258
+ "step": 363
2259
+ },
2260
+ {
2261
+ "epoch": 0.11,
2262
+ "learning_rate": 0.0002,
2263
+ "loss": 0.8927,
2264
+ "step": 364
2265
+ },
2266
+ {
2267
+ "epoch": 0.11,
2268
+ "learning_rate": 0.0002,
2269
+ "loss": 0.8619,
2270
+ "step": 365
2271
+ },
2272
+ {
2273
+ "epoch": 0.11,
2274
+ "learning_rate": 0.0002,
2275
+ "loss": 1.1063,
2276
+ "step": 366
2277
+ },
2278
+ {
2279
+ "epoch": 0.11,
2280
+ "learning_rate": 0.0002,
2281
+ "loss": 0.7939,
2282
+ "step": 367
2283
+ },
2284
+ {
2285
+ "epoch": 0.11,
2286
+ "learning_rate": 0.0002,
2287
+ "loss": 1.0938,
2288
+ "step": 368
2289
+ },
2290
+ {
2291
+ "epoch": 0.11,
2292
+ "learning_rate": 0.0002,
2293
+ "loss": 0.7703,
2294
+ "step": 369
2295
+ },
2296
+ {
2297
+ "epoch": 0.11,
2298
+ "learning_rate": 0.0002,
2299
+ "loss": 0.8846,
2300
+ "step": 370
2301
+ },
2302
+ {
2303
+ "epoch": 0.11,
2304
+ "learning_rate": 0.0002,
2305
+ "loss": 1.1139,
2306
+ "step": 371
2307
+ },
2308
+ {
2309
+ "epoch": 0.11,
2310
+ "learning_rate": 0.0002,
2311
+ "loss": 0.871,
2312
+ "step": 372
2313
+ },
2314
+ {
2315
+ "epoch": 0.12,
2316
+ "learning_rate": 0.0002,
2317
+ "loss": 1.0847,
2318
+ "step": 373
2319
+ },
2320
+ {
2321
+ "epoch": 0.12,
2322
+ "learning_rate": 0.0002,
2323
+ "loss": 0.9088,
2324
+ "step": 374
2325
+ },
2326
+ {
2327
+ "epoch": 0.12,
2328
+ "learning_rate": 0.0002,
2329
+ "loss": 0.8681,
2330
+ "step": 375
2331
+ },
2332
+ {
2333
+ "epoch": 0.12,
2334
+ "learning_rate": 0.0002,
2335
+ "loss": 0.9996,
2336
+ "step": 376
2337
+ },
2338
+ {
2339
+ "epoch": 0.12,
2340
+ "learning_rate": 0.0002,
2341
+ "loss": 0.8648,
2342
+ "step": 377
2343
+ },
2344
+ {
2345
+ "epoch": 0.12,
2346
+ "learning_rate": 0.0002,
2347
+ "loss": 0.85,
2348
+ "step": 378
2349
+ },
2350
+ {
2351
+ "epoch": 0.12,
2352
+ "learning_rate": 0.0002,
2353
+ "loss": 0.8583,
2354
+ "step": 379
2355
+ },
2356
+ {
2357
+ "epoch": 0.12,
2358
+ "learning_rate": 0.0002,
2359
+ "loss": 1.2312,
2360
+ "step": 380
2361
+ },
2362
+ {
2363
+ "epoch": 0.12,
2364
+ "learning_rate": 0.0002,
2365
+ "loss": 0.9006,
2366
+ "step": 381
2367
+ },
2368
+ {
2369
+ "epoch": 0.12,
2370
+ "learning_rate": 0.0002,
2371
+ "loss": 0.8996,
2372
+ "step": 382
2373
+ },
2374
+ {
2375
+ "epoch": 0.12,
2376
+ "learning_rate": 0.0002,
2377
+ "loss": 1.1244,
2378
+ "step": 383
2379
+ },
2380
+ {
2381
+ "epoch": 0.12,
2382
+ "learning_rate": 0.0002,
2383
+ "loss": 0.9585,
2384
+ "step": 384
2385
+ },
2386
+ {
2387
+ "epoch": 0.12,
2388
+ "learning_rate": 0.0002,
2389
+ "loss": 0.7312,
2390
+ "step": 385
2391
+ },
2392
+ {
2393
+ "epoch": 0.12,
2394
+ "learning_rate": 0.0002,
2395
+ "loss": 1.0706,
2396
+ "step": 386
2397
+ },
2398
+ {
2399
+ "epoch": 0.12,
2400
+ "learning_rate": 0.0002,
2401
+ "loss": 1.1169,
2402
+ "step": 387
2403
+ },
2404
+ {
2405
+ "epoch": 0.12,
2406
+ "learning_rate": 0.0002,
2407
+ "loss": 0.8343,
2408
+ "step": 388
2409
+ },
2410
+ {
2411
+ "epoch": 0.12,
2412
+ "learning_rate": 0.0002,
2413
+ "loss": 0.9726,
2414
+ "step": 389
2415
+ },
2416
+ {
2417
+ "epoch": 0.12,
2418
+ "learning_rate": 0.0002,
2419
+ "loss": 0.9733,
2420
+ "step": 390
2421
+ },
2422
+ {
2423
+ "epoch": 0.12,
2424
+ "learning_rate": 0.0002,
2425
+ "loss": 0.9283,
2426
+ "step": 391
2427
+ },
2428
+ {
2429
+ "epoch": 0.12,
2430
+ "learning_rate": 0.0002,
2431
+ "loss": 1.0867,
2432
+ "step": 392
2433
+ },
2434
+ {
2435
+ "epoch": 0.12,
2436
+ "learning_rate": 0.0002,
2437
+ "loss": 0.8833,
2438
+ "step": 393
2439
+ },
2440
+ {
2441
+ "epoch": 0.12,
2442
+ "learning_rate": 0.0002,
2443
+ "loss": 1.1476,
2444
+ "step": 394
2445
+ },
2446
+ {
2447
+ "epoch": 0.12,
2448
+ "learning_rate": 0.0002,
2449
+ "loss": 0.9248,
2450
+ "step": 395
2451
+ },
2452
+ {
2453
+ "epoch": 0.12,
2454
+ "learning_rate": 0.0002,
2455
+ "loss": 0.9299,
2456
+ "step": 396
2457
+ },
2458
+ {
2459
+ "epoch": 0.12,
2460
+ "learning_rate": 0.0002,
2461
+ "loss": 0.7002,
2462
+ "step": 397
2463
+ },
2464
+ {
2465
+ "epoch": 0.12,
2466
+ "learning_rate": 0.0002,
2467
+ "loss": 0.9364,
2468
+ "step": 398
2469
+ },
2470
+ {
2471
+ "epoch": 0.12,
2472
+ "learning_rate": 0.0002,
2473
+ "loss": 0.9676,
2474
+ "step": 399
2475
+ },
2476
+ {
2477
+ "epoch": 0.12,
2478
+ "learning_rate": 0.0002,
2479
+ "loss": 1.0887,
2480
+ "step": 400
2481
+ },
2482
+ {
2483
+ "epoch": 0.12,
2484
+ "eval_loss": 0.9337849020957947,
2485
+ "eval_runtime": 90.0647,
2486
+ "eval_samples_per_second": 11.103,
2487
+ "eval_steps_per_second": 5.552,
2488
+ "step": 400
2489
+ },
2490
+ {
2491
+ "epoch": 0.12,
2492
+ "mmlu_eval_accuracy": 0.5981609425223685,
2493
+ "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
2494
+ "mmlu_eval_accuracy_anatomy": 0.5,
2495
+ "mmlu_eval_accuracy_astronomy": 0.6875,
2496
+ "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
2497
+ "mmlu_eval_accuracy_clinical_knowledge": 0.5862068965517241,
2498
+ "mmlu_eval_accuracy_college_biology": 0.625,
2499
+ "mmlu_eval_accuracy_college_chemistry": 0.25,
2500
+ "mmlu_eval_accuracy_college_computer_science": 0.5454545454545454,
2501
+ "mmlu_eval_accuracy_college_mathematics": 0.6363636363636364,
2502
+ "mmlu_eval_accuracy_college_medicine": 0.6363636363636364,
2503
+ "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
2504
+ "mmlu_eval_accuracy_computer_security": 0.5454545454545454,
2505
+ "mmlu_eval_accuracy_conceptual_physics": 0.5,
2506
+ "mmlu_eval_accuracy_econometrics": 0.4166666666666667,
2507
+ "mmlu_eval_accuracy_electrical_engineering": 0.6875,
2508
+ "mmlu_eval_accuracy_elementary_mathematics": 0.43902439024390244,
2509
+ "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
2510
+ "mmlu_eval_accuracy_global_facts": 0.4,
2511
+ "mmlu_eval_accuracy_high_school_biology": 0.5,
2512
+ "mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
2513
+ "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
2514
+ "mmlu_eval_accuracy_high_school_european_history": 0.7222222222222222,
2515
+ "mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
2516
+ "mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
2517
+ "mmlu_eval_accuracy_high_school_macroeconomics": 0.5813953488372093,
2518
+ "mmlu_eval_accuracy_high_school_mathematics": 0.3793103448275862,
2519
+ "mmlu_eval_accuracy_high_school_microeconomics": 0.6538461538461539,
2520
+ "mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
2521
+ "mmlu_eval_accuracy_high_school_psychology": 0.85,
2522
+ "mmlu_eval_accuracy_high_school_statistics": 0.43478260869565216,
2523
+ "mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
2524
+ "mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
2525
+ "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
2526
+ "mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
2527
+ "mmlu_eval_accuracy_international_law": 0.9230769230769231,
2528
+ "mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
2529
+ "mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
2530
+ "mmlu_eval_accuracy_machine_learning": 0.45454545454545453,
2531
+ "mmlu_eval_accuracy_management": 0.9090909090909091,
2532
+ "mmlu_eval_accuracy_marketing": 0.96,
2533
+ "mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
2534
+ "mmlu_eval_accuracy_miscellaneous": 0.7558139534883721,
2535
+ "mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
2536
+ "mmlu_eval_accuracy_moral_scenarios": 0.31,
2537
+ "mmlu_eval_accuracy_nutrition": 0.7575757575757576,
2538
+ "mmlu_eval_accuracy_philosophy": 0.7647058823529411,
2539
+ "mmlu_eval_accuracy_prehistory": 0.5142857142857142,
2540
+ "mmlu_eval_accuracy_professional_accounting": 0.5806451612903226,
2541
+ "mmlu_eval_accuracy_professional_law": 0.40588235294117647,
2542
+ "mmlu_eval_accuracy_professional_medicine": 0.6129032258064516,
2543
+ "mmlu_eval_accuracy_professional_psychology": 0.6231884057971014,
2544
+ "mmlu_eval_accuracy_public_relations": 0.5,
2545
+ "mmlu_eval_accuracy_security_studies": 0.6296296296296297,
2546
+ "mmlu_eval_accuracy_sociology": 0.8636363636363636,
2547
+ "mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
2548
+ "mmlu_eval_accuracy_virology": 0.5555555555555556,
2549
+ "mmlu_eval_accuracy_world_religions": 0.8421052631578947,
2550
+ "mmlu_loss": 1.4234340800940213,
2551
+ "step": 400
2552
+ },
2553
+ {
2554
+ "epoch": 0.12,
2555
+ "learning_rate": 0.0002,
2556
+ "loss": 1.0558,
2557
+ "step": 401
2558
+ },
2559
+ {
2560
+ "epoch": 0.12,
2561
+ "learning_rate": 0.0002,
2562
+ "loss": 0.8911,
2563
+ "step": 402
2564
+ },
2565
+ {
2566
+ "epoch": 0.12,
2567
+ "learning_rate": 0.0002,
2568
+ "loss": 1.1997,
2569
+ "step": 403
2570
+ },
2571
+ {
2572
+ "epoch": 0.12,
2573
+ "learning_rate": 0.0002,
2574
+ "loss": 1.0156,
2575
+ "step": 404
2576
+ },
2577
+ {
2578
+ "epoch": 0.12,
2579
+ "learning_rate": 0.0002,
2580
+ "loss": 0.9449,
2581
+ "step": 405
2582
+ },
2583
+ {
2584
+ "epoch": 0.13,
2585
+ "learning_rate": 0.0002,
2586
+ "loss": 1.0698,
2587
+ "step": 406
2588
+ },
2589
+ {
2590
+ "epoch": 0.13,
2591
+ "learning_rate": 0.0002,
2592
+ "loss": 1.0274,
2593
+ "step": 407
2594
+ },
2595
+ {
2596
+ "epoch": 0.13,
2597
+ "learning_rate": 0.0002,
2598
+ "loss": 0.9211,
2599
+ "step": 408
2600
+ },
2601
+ {
2602
+ "epoch": 0.13,
2603
+ "learning_rate": 0.0002,
2604
+ "loss": 0.8839,
2605
+ "step": 409
2606
+ },
2607
+ {
2608
+ "epoch": 0.13,
2609
+ "learning_rate": 0.0002,
2610
+ "loss": 1.0304,
2611
+ "step": 410
2612
+ },
2613
+ {
2614
+ "epoch": 0.13,
2615
+ "learning_rate": 0.0002,
2616
+ "loss": 1.0496,
2617
+ "step": 411
2618
+ },
2619
+ {
2620
+ "epoch": 0.13,
2621
+ "learning_rate": 0.0002,
2622
+ "loss": 0.9538,
2623
+ "step": 412
2624
+ },
2625
+ {
2626
+ "epoch": 0.13,
2627
+ "learning_rate": 0.0002,
2628
+ "loss": 0.9134,
2629
+ "step": 413
2630
+ },
2631
+ {
2632
+ "epoch": 0.13,
2633
+ "learning_rate": 0.0002,
2634
+ "loss": 0.8762,
2635
+ "step": 414
2636
+ },
2637
+ {
2638
+ "epoch": 0.13,
2639
+ "learning_rate": 0.0002,
2640
+ "loss": 1.0562,
2641
+ "step": 415
2642
+ },
2643
+ {
2644
+ "epoch": 0.13,
2645
+ "learning_rate": 0.0002,
2646
+ "loss": 0.9841,
2647
+ "step": 416
2648
+ },
2649
+ {
2650
+ "epoch": 0.13,
2651
+ "learning_rate": 0.0002,
2652
+ "loss": 0.8964,
2653
+ "step": 417
2654
+ },
2655
+ {
2656
+ "epoch": 0.13,
2657
+ "learning_rate": 0.0002,
2658
+ "loss": 0.8863,
2659
+ "step": 418
2660
+ },
2661
+ {
2662
+ "epoch": 0.13,
2663
+ "learning_rate": 0.0002,
2664
+ "loss": 0.854,
2665
+ "step": 419
2666
+ },
2667
+ {
2668
+ "epoch": 0.13,
2669
+ "learning_rate": 0.0002,
2670
+ "loss": 0.9016,
2671
+ "step": 420
2672
+ },
2673
+ {
2674
+ "epoch": 0.13,
2675
+ "learning_rate": 0.0002,
2676
+ "loss": 0.938,
2677
+ "step": 421
2678
+ },
2679
+ {
2680
+ "epoch": 0.13,
2681
+ "learning_rate": 0.0002,
2682
+ "loss": 0.9199,
2683
+ "step": 422
2684
+ },
2685
+ {
2686
+ "epoch": 0.13,
2687
+ "learning_rate": 0.0002,
2688
+ "loss": 0.9212,
2689
+ "step": 423
2690
+ },
2691
+ {
2692
+ "epoch": 0.13,
2693
+ "learning_rate": 0.0002,
2694
+ "loss": 0.8638,
2695
+ "step": 424
2696
+ },
2697
+ {
2698
+ "epoch": 0.13,
2699
+ "learning_rate": 0.0002,
2700
+ "loss": 1.0294,
2701
+ "step": 425
2702
+ },
2703
+ {
2704
+ "epoch": 0.13,
2705
+ "learning_rate": 0.0002,
2706
+ "loss": 0.9055,
2707
+ "step": 426
2708
+ },
2709
+ {
2710
+ "epoch": 0.13,
2711
+ "learning_rate": 0.0002,
2712
+ "loss": 0.9948,
2713
+ "step": 427
2714
+ },
2715
+ {
2716
+ "epoch": 0.13,
2717
+ "learning_rate": 0.0002,
2718
+ "loss": 1.0142,
2719
+ "step": 428
2720
+ },
2721
+ {
2722
+ "epoch": 0.13,
2723
+ "learning_rate": 0.0002,
2724
+ "loss": 0.8889,
2725
+ "step": 429
2726
+ },
2727
+ {
2728
+ "epoch": 0.13,
2729
+ "learning_rate": 0.0002,
2730
+ "loss": 0.8868,
2731
+ "step": 430
2732
+ },
2733
+ {
2734
+ "epoch": 0.13,
2735
+ "learning_rate": 0.0002,
2736
+ "loss": 0.9281,
2737
+ "step": 431
2738
+ },
2739
+ {
2740
+ "epoch": 0.13,
2741
+ "learning_rate": 0.0002,
2742
+ "loss": 1.1905,
2743
+ "step": 432
2744
+ },
2745
+ {
2746
+ "epoch": 0.13,
2747
+ "learning_rate": 0.0002,
2748
+ "loss": 1.0471,
2749
+ "step": 433
2750
+ },
2751
+ {
2752
+ "epoch": 0.13,
2753
+ "learning_rate": 0.0002,
2754
+ "loss": 0.8069,
2755
+ "step": 434
2756
+ },
2757
+ {
2758
+ "epoch": 0.13,
2759
+ "learning_rate": 0.0002,
2760
+ "loss": 0.9336,
2761
+ "step": 435
2762
+ },
2763
+ {
2764
+ "epoch": 0.13,
2765
+ "learning_rate": 0.0002,
2766
+ "loss": 0.9766,
2767
+ "step": 436
2768
+ },
2769
+ {
2770
+ "epoch": 0.13,
2771
+ "learning_rate": 0.0002,
2772
+ "loss": 0.9063,
2773
+ "step": 437
2774
+ },
2775
+ {
2776
+ "epoch": 0.14,
2777
+ "learning_rate": 0.0002,
2778
+ "loss": 0.8862,
2779
+ "step": 438
2780
+ },
2781
+ {
2782
+ "epoch": 0.14,
2783
+ "learning_rate": 0.0002,
2784
+ "loss": 0.9648,
2785
+ "step": 439
2786
+ },
2787
+ {
2788
+ "epoch": 0.14,
2789
+ "learning_rate": 0.0002,
2790
+ "loss": 0.9775,
2791
+ "step": 440
2792
+ },
2793
+ {
2794
+ "epoch": 0.14,
2795
+ "learning_rate": 0.0002,
2796
+ "loss": 0.9181,
2797
+ "step": 441
2798
+ },
2799
+ {
2800
+ "epoch": 0.14,
2801
+ "learning_rate": 0.0002,
2802
+ "loss": 1.0634,
2803
+ "step": 442
2804
+ },
2805
+ {
2806
+ "epoch": 0.14,
2807
+ "learning_rate": 0.0002,
2808
+ "loss": 0.9453,
2809
+ "step": 443
2810
+ },
2811
+ {
2812
+ "epoch": 0.14,
2813
+ "learning_rate": 0.0002,
2814
+ "loss": 0.7123,
2815
+ "step": 444
2816
+ },
2817
+ {
2818
+ "epoch": 0.14,
2819
+ "learning_rate": 0.0002,
2820
+ "loss": 1.0168,
2821
+ "step": 445
2822
+ },
2823
+ {
2824
+ "epoch": 0.14,
2825
+ "learning_rate": 0.0002,
2826
+ "loss": 0.8962,
2827
+ "step": 446
2828
+ },
2829
+ {
2830
+ "epoch": 0.14,
2831
+ "learning_rate": 0.0002,
2832
+ "loss": 1.038,
2833
+ "step": 447
2834
+ },
2835
+ {
2836
+ "epoch": 0.14,
2837
+ "learning_rate": 0.0002,
2838
+ "loss": 0.8036,
2839
+ "step": 448
2840
+ },
2841
+ {
2842
+ "epoch": 0.14,
2843
+ "learning_rate": 0.0002,
2844
+ "loss": 0.9838,
2845
+ "step": 449
2846
+ },
2847
+ {
2848
+ "epoch": 0.14,
2849
+ "learning_rate": 0.0002,
2850
+ "loss": 0.6722,
2851
+ "step": 450
2852
+ },
2853
+ {
2854
+ "epoch": 0.14,
2855
+ "learning_rate": 0.0002,
2856
+ "loss": 0.9898,
2857
+ "step": 451
2858
+ },
2859
+ {
2860
+ "epoch": 0.14,
2861
+ "learning_rate": 0.0002,
2862
+ "loss": 1.2471,
2863
+ "step": 452
2864
+ },
2865
+ {
2866
+ "epoch": 0.14,
2867
+ "learning_rate": 0.0002,
2868
+ "loss": 0.9111,
2869
+ "step": 453
2870
+ },
2871
+ {
2872
+ "epoch": 0.14,
2873
+ "learning_rate": 0.0002,
2874
+ "loss": 1.1423,
2875
+ "step": 454
2876
+ },
2877
+ {
2878
+ "epoch": 0.14,
2879
+ "learning_rate": 0.0002,
2880
+ "loss": 0.6952,
2881
+ "step": 455
2882
+ },
2883
+ {
2884
+ "epoch": 0.14,
2885
+ "learning_rate": 0.0002,
2886
+ "loss": 1.0062,
2887
+ "step": 456
2888
+ },
2889
+ {
2890
+ "epoch": 0.14,
2891
+ "learning_rate": 0.0002,
2892
+ "loss": 0.9,
2893
+ "step": 457
2894
+ },
2895
+ {
2896
+ "epoch": 0.14,
2897
+ "learning_rate": 0.0002,
2898
+ "loss": 0.9786,
2899
+ "step": 458
2900
+ },
2901
+ {
2902
+ "epoch": 0.14,
2903
+ "learning_rate": 0.0002,
2904
+ "loss": 1.0072,
2905
+ "step": 459
2906
+ },
2907
+ {
2908
+ "epoch": 0.14,
2909
+ "learning_rate": 0.0002,
2910
+ "loss": 1.0386,
2911
+ "step": 460
2912
+ },
2913
+ {
2914
+ "epoch": 0.14,
2915
+ "learning_rate": 0.0002,
2916
+ "loss": 1.013,
2917
+ "step": 461
2918
+ },
2919
+ {
2920
+ "epoch": 0.14,
2921
+ "learning_rate": 0.0002,
2922
+ "loss": 0.9622,
2923
+ "step": 462
2924
+ },
2925
+ {
2926
+ "epoch": 0.14,
2927
+ "learning_rate": 0.0002,
2928
+ "loss": 0.9577,
2929
+ "step": 463
2930
+ },
2931
+ {
2932
+ "epoch": 0.14,
2933
+ "learning_rate": 0.0002,
2934
+ "loss": 0.9358,
2935
+ "step": 464
2936
+ },
2937
+ {
2938
+ "epoch": 0.14,
2939
+ "learning_rate": 0.0002,
2940
+ "loss": 0.853,
2941
+ "step": 465
2942
+ },
2943
+ {
2944
+ "epoch": 0.14,
2945
+ "learning_rate": 0.0002,
2946
+ "loss": 1.1448,
2947
+ "step": 466
2948
+ },
2949
+ {
2950
+ "epoch": 0.14,
2951
+ "learning_rate": 0.0002,
2952
+ "loss": 1.0459,
2953
+ "step": 467
2954
+ },
2955
+ {
2956
+ "epoch": 0.14,
2957
+ "learning_rate": 0.0002,
2958
+ "loss": 0.9386,
2959
+ "step": 468
2960
+ },
2961
+ {
2962
+ "epoch": 0.14,
2963
+ "learning_rate": 0.0002,
2964
+ "loss": 0.98,
2965
+ "step": 469
2966
+ },
2967
+ {
2968
+ "epoch": 0.15,
2969
+ "learning_rate": 0.0002,
2970
+ "loss": 0.995,
2971
+ "step": 470
2972
+ },
2973
+ {
2974
+ "epoch": 0.15,
2975
+ "learning_rate": 0.0002,
2976
+ "loss": 0.9831,
2977
+ "step": 471
2978
+ },
2979
+ {
2980
+ "epoch": 0.15,
2981
+ "learning_rate": 0.0002,
2982
+ "loss": 1.0843,
2983
+ "step": 472
2984
+ },
2985
+ {
2986
+ "epoch": 0.15,
2987
+ "learning_rate": 0.0002,
2988
+ "loss": 0.9832,
2989
+ "step": 473
2990
+ },
2991
+ {
2992
+ "epoch": 0.15,
2993
+ "learning_rate": 0.0002,
2994
+ "loss": 0.9159,
2995
+ "step": 474
2996
+ },
2997
+ {
2998
+ "epoch": 0.15,
2999
+ "learning_rate": 0.0002,
3000
+ "loss": 0.8165,
3001
+ "step": 475
3002
+ },
3003
+ {
3004
+ "epoch": 0.15,
3005
+ "learning_rate": 0.0002,
3006
+ "loss": 0.8995,
3007
+ "step": 476
3008
+ },
3009
+ {
3010
+ "epoch": 0.15,
3011
+ "learning_rate": 0.0002,
3012
+ "loss": 0.9109,
3013
+ "step": 477
3014
+ },
3015
+ {
3016
+ "epoch": 0.15,
3017
+ "learning_rate": 0.0002,
3018
+ "loss": 0.9167,
3019
+ "step": 478
3020
+ },
3021
+ {
3022
+ "epoch": 0.15,
3023
+ "learning_rate": 0.0002,
3024
+ "loss": 0.8154,
3025
+ "step": 479
3026
+ },
3027
+ {
3028
+ "epoch": 0.15,
3029
+ "learning_rate": 0.0002,
3030
+ "loss": 1.0069,
3031
+ "step": 480
3032
+ },
3033
+ {
3034
+ "epoch": 0.15,
3035
+ "learning_rate": 0.0002,
3036
+ "loss": 0.9082,
3037
+ "step": 481
3038
+ },
3039
+ {
3040
+ "epoch": 0.15,
3041
+ "learning_rate": 0.0002,
3042
+ "loss": 0.9815,
3043
+ "step": 482
3044
+ },
3045
+ {
3046
+ "epoch": 0.15,
3047
+ "learning_rate": 0.0002,
3048
+ "loss": 0.8454,
3049
+ "step": 483
3050
+ },
3051
+ {
3052
+ "epoch": 0.15,
3053
+ "learning_rate": 0.0002,
3054
+ "loss": 0.8938,
3055
+ "step": 484
3056
+ },
3057
+ {
3058
+ "epoch": 0.15,
3059
+ "learning_rate": 0.0002,
3060
+ "loss": 0.9268,
3061
+ "step": 485
3062
+ },
3063
+ {
3064
+ "epoch": 0.15,
3065
+ "learning_rate": 0.0002,
3066
+ "loss": 0.8692,
3067
+ "step": 486
3068
+ },
3069
+ {
3070
+ "epoch": 0.15,
3071
+ "learning_rate": 0.0002,
3072
+ "loss": 1.141,
3073
+ "step": 487
3074
+ },
3075
+ {
3076
+ "epoch": 0.15,
3077
+ "learning_rate": 0.0002,
3078
+ "loss": 0.8405,
3079
+ "step": 488
3080
+ },
3081
+ {
3082
+ "epoch": 0.15,
3083
+ "learning_rate": 0.0002,
3084
+ "loss": 0.8733,
3085
+ "step": 489
3086
+ },
3087
+ {
3088
+ "epoch": 0.15,
3089
+ "learning_rate": 0.0002,
3090
+ "loss": 0.9348,
3091
+ "step": 490
3092
+ },
3093
+ {
3094
+ "epoch": 0.15,
3095
+ "learning_rate": 0.0002,
3096
+ "loss": 0.9177,
3097
+ "step": 491
3098
+ },
3099
+ {
3100
+ "epoch": 0.15,
3101
+ "learning_rate": 0.0002,
3102
+ "loss": 1.007,
3103
+ "step": 492
3104
+ },
3105
+ {
3106
+ "epoch": 0.15,
3107
+ "learning_rate": 0.0002,
3108
+ "loss": 1.1407,
3109
+ "step": 493
3110
+ },
3111
+ {
3112
+ "epoch": 0.15,
3113
+ "learning_rate": 0.0002,
3114
+ "loss": 0.9823,
3115
+ "step": 494
3116
+ },
3117
+ {
3118
+ "epoch": 0.15,
3119
+ "learning_rate": 0.0002,
3120
+ "loss": 0.9559,
3121
+ "step": 495
3122
+ },
3123
+ {
3124
+ "epoch": 0.15,
3125
+ "learning_rate": 0.0002,
3126
+ "loss": 0.8707,
3127
+ "step": 496
3128
+ },
3129
+ {
3130
+ "epoch": 0.15,
3131
+ "learning_rate": 0.0002,
3132
+ "loss": 1.0767,
3133
+ "step": 497
3134
+ },
3135
+ {
3136
+ "epoch": 0.15,
3137
+ "learning_rate": 0.0002,
3138
+ "loss": 0.8656,
3139
+ "step": 498
3140
+ },
3141
+ {
3142
+ "epoch": 0.15,
3143
+ "learning_rate": 0.0002,
3144
+ "loss": 0.9875,
3145
+ "step": 499
3146
+ },
3147
+ {
3148
+ "epoch": 0.15,
3149
+ "learning_rate": 0.0002,
3150
+ "loss": 0.8553,
3151
+ "step": 500
3152
+ },
3153
+ {
3154
+ "epoch": 0.15,
3155
+ "learning_rate": 0.0002,
3156
+ "loss": 1.2432,
3157
+ "step": 501
3158
+ },
3159
+ {
3160
+ "epoch": 0.15,
3161
+ "learning_rate": 0.0002,
3162
+ "loss": 0.9875,
3163
+ "step": 502
3164
+ },
3165
+ {
3166
+ "epoch": 0.16,
3167
+ "learning_rate": 0.0002,
3168
+ "loss": 1.0564,
3169
+ "step": 503
3170
+ },
3171
+ {
3172
+ "epoch": 0.16,
3173
+ "learning_rate": 0.0002,
3174
+ "loss": 1.0728,
3175
+ "step": 504
3176
+ },
3177
+ {
3178
+ "epoch": 0.16,
3179
+ "learning_rate": 0.0002,
3180
+ "loss": 0.8802,
3181
+ "step": 505
3182
+ },
3183
+ {
3184
+ "epoch": 0.16,
3185
+ "learning_rate": 0.0002,
3186
+ "loss": 1.0073,
3187
+ "step": 506
3188
+ },
3189
+ {
3190
+ "epoch": 0.16,
3191
+ "learning_rate": 0.0002,
3192
+ "loss": 1.0117,
3193
+ "step": 507
3194
+ },
3195
+ {
3196
+ "epoch": 0.16,
3197
+ "learning_rate": 0.0002,
3198
+ "loss": 0.9091,
3199
+ "step": 508
3200
+ },
3201
+ {
3202
+ "epoch": 0.16,
3203
+ "learning_rate": 0.0002,
3204
+ "loss": 1.0593,
3205
+ "step": 509
3206
+ },
3207
+ {
3208
+ "epoch": 0.16,
3209
+ "learning_rate": 0.0002,
3210
+ "loss": 0.7527,
3211
+ "step": 510
3212
+ },
3213
+ {
3214
+ "epoch": 0.16,
3215
+ "learning_rate": 0.0002,
3216
+ "loss": 0.884,
3217
+ "step": 511
3218
+ },
3219
+ {
3220
+ "epoch": 0.16,
3221
+ "learning_rate": 0.0002,
3222
+ "loss": 0.6971,
3223
+ "step": 512
3224
+ },
3225
+ {
3226
+ "epoch": 0.16,
3227
+ "learning_rate": 0.0002,
3228
+ "loss": 0.8787,
3229
+ "step": 513
3230
+ },
3231
+ {
3232
+ "epoch": 0.16,
3233
+ "learning_rate": 0.0002,
3234
+ "loss": 0.8654,
3235
+ "step": 514
3236
+ },
3237
+ {
3238
+ "epoch": 0.16,
3239
+ "learning_rate": 0.0002,
3240
+ "loss": 1.0796,
3241
+ "step": 515
3242
+ },
3243
+ {
3244
+ "epoch": 0.16,
3245
+ "learning_rate": 0.0002,
3246
+ "loss": 0.9112,
3247
+ "step": 516
3248
+ },
3249
+ {
3250
+ "epoch": 0.16,
3251
+ "learning_rate": 0.0002,
3252
+ "loss": 1.0223,
3253
+ "step": 517
3254
+ },
3255
+ {
3256
+ "epoch": 0.16,
3257
+ "learning_rate": 0.0002,
3258
+ "loss": 1.1011,
3259
+ "step": 518
3260
+ },
3261
+ {
3262
+ "epoch": 0.16,
3263
+ "learning_rate": 0.0002,
3264
+ "loss": 0.9859,
3265
+ "step": 519
3266
+ },
3267
+ {
3268
+ "epoch": 0.16,
3269
+ "learning_rate": 0.0002,
3270
+ "loss": 0.9615,
3271
+ "step": 520
3272
+ },
3273
+ {
3274
+ "epoch": 0.16,
3275
+ "learning_rate": 0.0002,
3276
+ "loss": 0.9893,
3277
+ "step": 521
3278
+ },
3279
+ {
3280
+ "epoch": 0.16,
3281
+ "learning_rate": 0.0002,
3282
+ "loss": 0.9579,
3283
+ "step": 522
3284
+ },
3285
+ {
3286
+ "epoch": 0.16,
3287
+ "learning_rate": 0.0002,
3288
+ "loss": 0.7488,
3289
+ "step": 523
3290
+ },
3291
+ {
3292
+ "epoch": 0.16,
3293
+ "learning_rate": 0.0002,
3294
+ "loss": 0.8454,
3295
+ "step": 524
3296
+ },
3297
+ {
3298
+ "epoch": 0.16,
3299
+ "learning_rate": 0.0002,
3300
+ "loss": 0.7769,
3301
+ "step": 525
3302
+ },
3303
+ {
3304
+ "epoch": 0.16,
3305
+ "learning_rate": 0.0002,
3306
+ "loss": 1.0176,
3307
+ "step": 526
3308
+ },
3309
+ {
3310
+ "epoch": 0.16,
3311
+ "learning_rate": 0.0002,
3312
+ "loss": 0.9641,
3313
+ "step": 527
3314
+ },
3315
+ {
3316
+ "epoch": 0.16,
3317
+ "learning_rate": 0.0002,
3318
+ "loss": 0.9485,
3319
+ "step": 528
3320
+ },
3321
+ {
3322
+ "epoch": 0.16,
3323
+ "learning_rate": 0.0002,
3324
+ "loss": 1.125,
3325
+ "step": 529
3326
+ },
3327
+ {
3328
+ "epoch": 0.16,
3329
+ "learning_rate": 0.0002,
3330
+ "loss": 1.1411,
3331
+ "step": 530
3332
+ },
3333
+ {
3334
+ "epoch": 0.16,
3335
+ "learning_rate": 0.0002,
3336
+ "loss": 0.8413,
3337
+ "step": 531
3338
+ },
3339
+ {
3340
+ "epoch": 0.16,
3341
+ "learning_rate": 0.0002,
3342
+ "loss": 1.1159,
3343
+ "step": 532
3344
+ },
3345
+ {
3346
+ "epoch": 0.16,
3347
+ "learning_rate": 0.0002,
3348
+ "loss": 1.0399,
3349
+ "step": 533
3350
+ },
3351
+ {
3352
+ "epoch": 0.16,
3353
+ "learning_rate": 0.0002,
3354
+ "loss": 0.9691,
3355
+ "step": 534
3356
+ },
3357
+ {
3358
+ "epoch": 0.17,
3359
+ "learning_rate": 0.0002,
3360
+ "loss": 1.038,
3361
+ "step": 535
3362
+ },
3363
+ {
3364
+ "epoch": 0.17,
3365
+ "learning_rate": 0.0002,
3366
+ "loss": 0.7182,
3367
+ "step": 536
3368
+ },
3369
+ {
3370
+ "epoch": 0.17,
3371
+ "learning_rate": 0.0002,
3372
+ "loss": 0.9709,
3373
+ "step": 537
3374
+ },
3375
+ {
3376
+ "epoch": 0.17,
3377
+ "learning_rate": 0.0002,
3378
+ "loss": 0.9577,
3379
+ "step": 538
3380
+ },
3381
+ {
3382
+ "epoch": 0.17,
3383
+ "learning_rate": 0.0002,
3384
+ "loss": 0.9499,
3385
+ "step": 539
3386
+ },
3387
+ {
3388
+ "epoch": 0.17,
3389
+ "learning_rate": 0.0002,
3390
+ "loss": 0.9988,
3391
+ "step": 540
3392
+ },
3393
+ {
3394
+ "epoch": 0.17,
3395
+ "learning_rate": 0.0002,
3396
+ "loss": 0.9525,
3397
+ "step": 541
3398
+ },
3399
+ {
3400
+ "epoch": 0.17,
3401
+ "learning_rate": 0.0002,
3402
+ "loss": 0.8858,
3403
+ "step": 542
3404
+ },
3405
+ {
3406
+ "epoch": 0.17,
3407
+ "learning_rate": 0.0002,
3408
+ "loss": 0.9578,
3409
+ "step": 543
3410
+ },
3411
+ {
3412
+ "epoch": 0.17,
3413
+ "learning_rate": 0.0002,
3414
+ "loss": 1.0587,
3415
+ "step": 544
3416
+ },
3417
+ {
3418
+ "epoch": 0.17,
3419
+ "learning_rate": 0.0002,
3420
+ "loss": 1.0067,
3421
+ "step": 545
3422
+ },
3423
+ {
3424
+ "epoch": 0.17,
3425
+ "learning_rate": 0.0002,
3426
+ "loss": 0.955,
3427
+ "step": 546
3428
+ },
3429
+ {
3430
+ "epoch": 0.17,
3431
+ "learning_rate": 0.0002,
3432
+ "loss": 1.0419,
3433
+ "step": 547
3434
+ },
3435
+ {
3436
+ "epoch": 0.17,
3437
+ "learning_rate": 0.0002,
3438
+ "loss": 1.1421,
3439
+ "step": 548
3440
+ },
3441
+ {
3442
+ "epoch": 0.17,
3443
+ "learning_rate": 0.0002,
3444
+ "loss": 0.9016,
3445
+ "step": 549
3446
+ },
3447
+ {
3448
+ "epoch": 0.17,
3449
+ "learning_rate": 0.0002,
3450
+ "loss": 1.1797,
3451
+ "step": 550
3452
+ },
3453
+ {
3454
+ "epoch": 0.17,
3455
+ "learning_rate": 0.0002,
3456
+ "loss": 0.8918,
3457
+ "step": 551
3458
+ },
3459
+ {
3460
+ "epoch": 0.17,
3461
+ "learning_rate": 0.0002,
3462
+ "loss": 0.978,
3463
+ "step": 552
3464
+ },
3465
+ {
3466
+ "epoch": 0.17,
3467
+ "learning_rate": 0.0002,
3468
+ "loss": 0.901,
3469
+ "step": 553
3470
+ },
3471
+ {
3472
+ "epoch": 0.17,
3473
+ "learning_rate": 0.0002,
3474
+ "loss": 0.9897,
3475
+ "step": 554
3476
+ },
3477
+ {
3478
+ "epoch": 0.17,
3479
+ "learning_rate": 0.0002,
3480
+ "loss": 1.0817,
3481
+ "step": 555
3482
+ },
3483
+ {
3484
+ "epoch": 0.17,
3485
+ "learning_rate": 0.0002,
3486
+ "loss": 0.9277,
3487
+ "step": 556
3488
+ },
3489
+ {
3490
+ "epoch": 0.17,
3491
+ "learning_rate": 0.0002,
3492
+ "loss": 0.7438,
3493
+ "step": 557
3494
+ },
3495
+ {
3496
+ "epoch": 0.17,
3497
+ "learning_rate": 0.0002,
3498
+ "loss": 0.9909,
3499
+ "step": 558
3500
+ },
3501
+ {
3502
+ "epoch": 0.17,
3503
+ "learning_rate": 0.0002,
3504
+ "loss": 0.9218,
3505
+ "step": 559
3506
+ },
3507
+ {
3508
+ "epoch": 0.17,
3509
+ "learning_rate": 0.0002,
3510
+ "loss": 0.9851,
3511
+ "step": 560
3512
+ },
3513
+ {
3514
+ "epoch": 0.17,
3515
+ "learning_rate": 0.0002,
3516
+ "loss": 0.9524,
3517
+ "step": 561
3518
+ },
3519
+ {
3520
+ "epoch": 0.17,
3521
+ "learning_rate": 0.0002,
3522
+ "loss": 0.8747,
3523
+ "step": 562
3524
+ },
3525
+ {
3526
+ "epoch": 0.17,
3527
+ "learning_rate": 0.0002,
3528
+ "loss": 0.9278,
3529
+ "step": 563
3530
+ },
3531
+ {
3532
+ "epoch": 0.17,
3533
+ "learning_rate": 0.0002,
3534
+ "loss": 0.9328,
3535
+ "step": 564
3536
+ },
3537
+ {
3538
+ "epoch": 0.17,
3539
+ "learning_rate": 0.0002,
3540
+ "loss": 0.9843,
3541
+ "step": 565
3542
+ },
3543
+ {
3544
+ "epoch": 0.17,
3545
+ "learning_rate": 0.0002,
3546
+ "loss": 0.7991,
3547
+ "step": 566
3548
+ },
3549
+ {
3550
+ "epoch": 0.17,
3551
+ "learning_rate": 0.0002,
3552
+ "loss": 0.976,
3553
+ "step": 567
3554
+ },
3555
+ {
3556
+ "epoch": 0.18,
3557
+ "learning_rate": 0.0002,
3558
+ "loss": 0.8677,
3559
+ "step": 568
3560
+ },
3561
+ {
3562
+ "epoch": 0.18,
3563
+ "learning_rate": 0.0002,
3564
+ "loss": 0.965,
3565
+ "step": 569
3566
+ },
3567
+ {
3568
+ "epoch": 0.18,
3569
+ "learning_rate": 0.0002,
3570
+ "loss": 0.8707,
3571
+ "step": 570
3572
+ },
3573
+ {
3574
+ "epoch": 0.18,
3575
+ "learning_rate": 0.0002,
3576
+ "loss": 1.0468,
3577
+ "step": 571
3578
+ },
3579
+ {
3580
+ "epoch": 0.18,
3581
+ "learning_rate": 0.0002,
3582
+ "loss": 0.9707,
3583
+ "step": 572
3584
+ },
3585
+ {
3586
+ "epoch": 0.18,
3587
+ "learning_rate": 0.0002,
3588
+ "loss": 0.9433,
3589
+ "step": 573
3590
+ },
3591
+ {
3592
+ "epoch": 0.18,
3593
+ "learning_rate": 0.0002,
3594
+ "loss": 1.0574,
3595
+ "step": 574
3596
+ },
3597
+ {
3598
+ "epoch": 0.18,
3599
+ "learning_rate": 0.0002,
3600
+ "loss": 1.0994,
3601
+ "step": 575
3602
+ },
3603
+ {
3604
+ "epoch": 0.18,
3605
+ "learning_rate": 0.0002,
3606
+ "loss": 0.8638,
3607
+ "step": 576
3608
+ },
3609
+ {
3610
+ "epoch": 0.18,
3611
+ "learning_rate": 0.0002,
3612
+ "loss": 0.8006,
3613
+ "step": 577
3614
+ },
3615
+ {
3616
+ "epoch": 0.18,
3617
+ "learning_rate": 0.0002,
3618
+ "loss": 0.8519,
3619
+ "step": 578
3620
+ },
3621
+ {
3622
+ "epoch": 0.18,
3623
+ "learning_rate": 0.0002,
3624
+ "loss": 0.8249,
3625
+ "step": 579
3626
+ },
3627
+ {
3628
+ "epoch": 0.18,
3629
+ "learning_rate": 0.0002,
3630
+ "loss": 1.0455,
3631
+ "step": 580
3632
+ },
3633
+ {
3634
+ "epoch": 0.18,
3635
+ "learning_rate": 0.0002,
3636
+ "loss": 0.9342,
3637
+ "step": 581
3638
+ },
3639
+ {
3640
+ "epoch": 0.18,
3641
+ "learning_rate": 0.0002,
3642
+ "loss": 0.8356,
3643
+ "step": 582
3644
+ },
3645
+ {
3646
+ "epoch": 0.18,
3647
+ "learning_rate": 0.0002,
3648
+ "loss": 0.8214,
3649
+ "step": 583
3650
+ },
3651
+ {
3652
+ "epoch": 0.18,
3653
+ "learning_rate": 0.0002,
3654
+ "loss": 1.1784,
3655
+ "step": 584
3656
+ },
3657
+ {
3658
+ "epoch": 0.18,
3659
+ "learning_rate": 0.0002,
3660
+ "loss": 0.9906,
3661
+ "step": 585
3662
+ },
3663
+ {
3664
+ "epoch": 0.18,
3665
+ "learning_rate": 0.0002,
3666
+ "loss": 0.7492,
3667
+ "step": 586
3668
+ },
3669
+ {
3670
+ "epoch": 0.18,
3671
+ "learning_rate": 0.0002,
3672
+ "loss": 0.9585,
3673
+ "step": 587
3674
+ },
3675
+ {
3676
+ "epoch": 0.18,
3677
+ "learning_rate": 0.0002,
3678
+ "loss": 0.8765,
3679
+ "step": 588
3680
+ },
3681
+ {
3682
+ "epoch": 0.18,
3683
+ "learning_rate": 0.0002,
3684
+ "loss": 0.9779,
3685
+ "step": 589
3686
+ },
3687
+ {
3688
+ "epoch": 0.18,
3689
+ "learning_rate": 0.0002,
3690
+ "loss": 1.3717,
3691
+ "step": 590
3692
+ },
3693
+ {
3694
+ "epoch": 0.18,
3695
+ "learning_rate": 0.0002,
3696
+ "loss": 0.9579,
3697
+ "step": 591
3698
+ },
3699
+ {
3700
+ "epoch": 0.18,
3701
+ "learning_rate": 0.0002,
3702
+ "loss": 0.8574,
3703
+ "step": 592
3704
+ },
3705
+ {
3706
+ "epoch": 0.18,
3707
+ "learning_rate": 0.0002,
3708
+ "loss": 0.9953,
3709
+ "step": 593
3710
+ },
3711
+ {
3712
+ "epoch": 0.18,
3713
+ "learning_rate": 0.0002,
3714
+ "loss": 0.9189,
3715
+ "step": 594
3716
+ },
3717
+ {
3718
+ "epoch": 0.18,
3719
+ "learning_rate": 0.0002,
3720
+ "loss": 0.8172,
3721
+ "step": 595
3722
+ },
3723
+ {
3724
+ "epoch": 0.18,
3725
+ "learning_rate": 0.0002,
3726
+ "loss": 1.0521,
3727
+ "step": 596
3728
+ },
3729
+ {
3730
+ "epoch": 0.18,
3731
+ "learning_rate": 0.0002,
3732
+ "loss": 0.9594,
3733
+ "step": 597
3734
+ },
3735
+ {
3736
+ "epoch": 0.18,
3737
+ "learning_rate": 0.0002,
3738
+ "loss": 0.988,
3739
+ "step": 598
3740
+ },
3741
+ {
3742
+ "epoch": 0.18,
3743
+ "learning_rate": 0.0002,
3744
+ "loss": 1.1481,
3745
+ "step": 599
3746
+ },
3747
+ {
3748
+ "epoch": 0.19,
3749
+ "learning_rate": 0.0002,
3750
+ "loss": 0.9953,
3751
+ "step": 600
3752
+ },
3753
+ {
3754
+ "epoch": 0.19,
3755
+ "eval_loss": 0.9254454374313354,
3756
+ "eval_runtime": 89.8948,
3757
+ "eval_samples_per_second": 11.124,
3758
+ "eval_steps_per_second": 5.562,
3759
+ "step": 600
3760
+ },
3761
+ {
3762
+ "epoch": 0.19,
3763
+ "mmlu_eval_accuracy": 0.5822425888854337,
3764
+ "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
3765
+ "mmlu_eval_accuracy_anatomy": 0.5,
3766
+ "mmlu_eval_accuracy_astronomy": 0.6875,
3767
+ "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
3768
+ "mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
3769
+ "mmlu_eval_accuracy_college_biology": 0.625,
3770
+ "mmlu_eval_accuracy_college_chemistry": 0.25,
3771
+ "mmlu_eval_accuracy_college_computer_science": 0.18181818181818182,
3772
+ "mmlu_eval_accuracy_college_mathematics": 0.5454545454545454,
3773
+ "mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
3774
+ "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
3775
+ "mmlu_eval_accuracy_computer_security": 0.5454545454545454,
3776
+ "mmlu_eval_accuracy_conceptual_physics": 0.5,
3777
+ "mmlu_eval_accuracy_econometrics": 0.5,
3778
+ "mmlu_eval_accuracy_electrical_engineering": 0.5625,
3779
+ "mmlu_eval_accuracy_elementary_mathematics": 0.4878048780487805,
3780
+ "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
3781
+ "mmlu_eval_accuracy_global_facts": 0.3,
3782
+ "mmlu_eval_accuracy_high_school_biology": 0.53125,
3783
+ "mmlu_eval_accuracy_high_school_chemistry": 0.3181818181818182,
3784
+ "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
3785
+ "mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
3786
+ "mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
3787
+ "mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
3788
+ "mmlu_eval_accuracy_high_school_macroeconomics": 0.6046511627906976,
3789
+ "mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
3790
+ "mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
3791
+ "mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
3792
+ "mmlu_eval_accuracy_high_school_psychology": 0.85,
3793
+ "mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
3794
+ "mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
3795
+ "mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
3796
+ "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
3797
+ "mmlu_eval_accuracy_human_sexuality": 0.5,
3798
+ "mmlu_eval_accuracy_international_law": 0.9230769230769231,
3799
+ "mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
3800
+ "mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
3801
+ "mmlu_eval_accuracy_machine_learning": 0.5454545454545454,
3802
+ "mmlu_eval_accuracy_management": 0.9090909090909091,
3803
+ "mmlu_eval_accuracy_marketing": 0.92,
3804
+ "mmlu_eval_accuracy_medical_genetics": 1.0,
3805
+ "mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
3806
+ "mmlu_eval_accuracy_moral_disputes": 0.5,
3807
+ "mmlu_eval_accuracy_moral_scenarios": 0.37,
3808
+ "mmlu_eval_accuracy_nutrition": 0.6666666666666666,
3809
+ "mmlu_eval_accuracy_philosophy": 0.7352941176470589,
3810
+ "mmlu_eval_accuracy_prehistory": 0.5428571428571428,
3811
+ "mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
3812
+ "mmlu_eval_accuracy_professional_law": 0.3764705882352941,
3813
+ "mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
3814
+ "mmlu_eval_accuracy_professional_psychology": 0.6521739130434783,
3815
+ "mmlu_eval_accuracy_public_relations": 0.5,
3816
+ "mmlu_eval_accuracy_security_studies": 0.6296296296296297,
3817
+ "mmlu_eval_accuracy_sociology": 0.9090909090909091,
3818
+ "mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
3819
+ "mmlu_eval_accuracy_virology": 0.5555555555555556,
3820
+ "mmlu_eval_accuracy_world_religions": 0.8421052631578947,
3821
+ "mmlu_loss": 1.3702809761227266,
3822
+ "step": 600
3823
+ }
3824
+ ],
3825
+ "logging_steps": 1,
3826
+ "max_steps": 9723,
3827
+ "num_train_epochs": 3,
3828
+ "save_steps": 200,
3829
+ "total_flos": 6.563255848973107e+16,
3830
+ "trial_name": null,
3831
+ "trial_params": null
3832
+ }
checkpoint-600/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7a6b2b240ec721eec6ccab5248f6a812846a8257e0a9d8205b467e6601dd27e
3
+ size 6392