Hugging Face's logo

pharaouk
/
code-dex-1

Model card Files Community
Farouk commited on
Commit
12e9e28
·
1 Parent(s): e28fd11

Training in progress, step 600

Browse files
Files changed (16) hide show
  1. adapter_model.bin +1 -1
  2. checkpoint-400/adapter_model/adapter_model/README.md +20 -0
  3. checkpoint-400/adapter_model/adapter_model/adapter_config.json +26 -0
  4. checkpoint-400/adapter_model/adapter_model/adapter_model.bin +3 -0
  5. checkpoint-600/README.md +20 -0
  6. checkpoint-600/adapter_config.json +26 -0
  7. checkpoint-600/adapter_model.bin +3 -0
  8. checkpoint-600/added_tokens.json +3 -0
  9. checkpoint-600/optimizer.pt +3 -0
  10. checkpoint-600/rng_state.pth +3 -0
  11. checkpoint-600/scheduler.pt +3 -0
  12. checkpoint-600/special_tokens_map.json +6 -0
  13. checkpoint-600/tokenizer.model +3 -0
  14. checkpoint-600/tokenizer_config.json +37 -0
  15. checkpoint-600/trainer_state.json +3829 -0
  16. checkpoint-600/training_args.bin +3 -0
adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ab953b22d5f8efb3e7e86af263d15edd6f0cb816bed1b4cba209c25bc687d5b5
3
  size 871609293
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e000ae273775396215b4c2b8d647490934fbe43156330d8630253bc0cf696822
3
  size 871609293
checkpoint-400/adapter_model/adapter_model/README.md ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - load_in_8bit: False
9
+ - load_in_4bit: True
10
+ - llm_int8_threshold: 6.0
11
+ - llm_int8_skip_modules: None
12
+ - llm_int8_enable_fp32_cpu_offload: False
13
+ - llm_int8_has_fp16_weight: False
14
+ - bnb_4bit_quant_type: nf4
15
+ - bnb_4bit_use_double_quant: True
16
+ - bnb_4bit_compute_dtype: bfloat16
17
+ ### Framework versions
18
+
19
+
20
+ - PEFT 0.4.0
checkpoint-400/adapter_model/adapter_model/adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "alpindale/CodeLlama-34B-hf",
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
+ "q_proj",
18
+ "o_proj",
19
+ "k_proj",
20
+ "up_proj",
21
+ "down_proj",
22
+ "v_proj",
23
+ "gate_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
checkpoint-400/adapter_model/adapter_model/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab953b22d5f8efb3e7e86af263d15edd6f0cb816bed1b4cba209c25bc687d5b5
3
+ size 871609293
checkpoint-600/README.md ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - load_in_8bit: False
9
+ - load_in_4bit: True
10
+ - llm_int8_threshold: 6.0
11
+ - llm_int8_skip_modules: None
12
+ - llm_int8_enable_fp32_cpu_offload: False
13
+ - llm_int8_has_fp16_weight: False
14
+ - bnb_4bit_quant_type: nf4
15
+ - bnb_4bit_use_double_quant: True
16
+ - bnb_4bit_compute_dtype: bfloat16
17
+ ### Framework versions
18
+
19
+
20
+ - 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": "alpindale/CodeLlama-34B-hf",
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
+ "q_proj",
18
+ "o_proj",
19
+ "k_proj",
20
+ "up_proj",
21
+ "down_proj",
22
+ "v_proj",
23
+ "gate_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:e000ae273775396215b4c2b8d647490934fbe43156330d8630253bc0cf696822
3
+ size 871609293
checkpoint-600/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "[PAD]": 32000
3
+ }
checkpoint-600/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df789276400ab4313e4ca699e4165441f7bdaf5cdbbf5eed27a1b150f2658ae8
3
+ size 3485881117
checkpoint-600/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fffba8f7a22edd6afc5fcb3826998e70ed1cf4f1e856e3292b195eb164d7b54a
3
+ size 14511
checkpoint-600/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c1a2ac1c11599601eeac95feb1dbfd49ec5c625e61dcce18b3f094491f9cf2d
3
+ size 627
checkpoint-600/special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "pad_token": "[PAD]",
5
+ "unk_token": "<unk>"
6
+ }
checkpoint-600/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-600/tokenizer_config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "bos_token": {
5
+ "__type": "AddedToken",
6
+ "content": "<s>",
7
+ "lstrip": false,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "clean_up_tokenization_spaces": false,
13
+ "eos_token": {
14
+ "__type": "AddedToken",
15
+ "content": "</s>",
16
+ "lstrip": false,
17
+ "normalized": true,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "legacy": null,
22
+ "model_max_length": 1000000000000000019884624838656,
23
+ "pad_token": null,
24
+ "padding_side": "right",
25
+ "sp_model_kwargs": {},
26
+ "spaces_between_special_tokens": false,
27
+ "tokenizer_class": "LlamaTokenizer",
28
+ "unk_token": {
29
+ "__type": "AddedToken",
30
+ "content": "<unk>",
31
+ "lstrip": false,
32
+ "normalized": true,
33
+ "rstrip": false,
34
+ "single_word": false
35
+ },
36
+ "use_default_system_prompt": true
37
+ }
checkpoint-600/trainer_state.json ADDED
@@ -0,0 +1,3829 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.5618544220924377,
3
+ "best_model_checkpoint": "./output_v2/34bCodellama_CodeLlama-34B-hf_codellama_blob_1/checkpoint-600",
4
+ "epoch": 0.8424008424008425,
5
+ "global_step": 600,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.0,
12
+ "learning_rate": 0.0002,
13
+ "loss": 1.5798,
14
+ "step": 1
15
+ },
16
+ {
17
+ "epoch": 0.0,
18
+ "learning_rate": 0.0002,
19
+ "loss": 1.6599,
20
+ "step": 2
21
+ },
22
+ {
23
+ "epoch": 0.0,
24
+ "learning_rate": 0.0002,
25
+ "loss": 1.6807,
26
+ "step": 3
27
+ },
28
+ {
29
+ "epoch": 0.01,
30
+ "learning_rate": 0.0002,
31
+ "loss": 1.4016,
32
+ "step": 4
33
+ },
34
+ {
35
+ "epoch": 0.01,
36
+ "learning_rate": 0.0002,
37
+ "loss": 1.2944,
38
+ "step": 5
39
+ },
40
+ {
41
+ "epoch": 0.01,
42
+ "learning_rate": 0.0002,
43
+ "loss": 1.5477,
44
+ "step": 6
45
+ },
46
+ {
47
+ "epoch": 0.01,
48
+ "learning_rate": 0.0002,
49
+ "loss": 1.1555,
50
+ "step": 7
51
+ },
52
+ {
53
+ "epoch": 0.01,
54
+ "learning_rate": 0.0002,
55
+ "loss": 0.8678,
56
+ "step": 8
57
+ },
58
+ {
59
+ "epoch": 0.01,
60
+ "learning_rate": 0.0002,
61
+ "loss": 1.1943,
62
+ "step": 9
63
+ },
64
+ {
65
+ "epoch": 0.01,
66
+ "learning_rate": 0.0002,
67
+ "loss": 0.8046,
68
+ "step": 10
69
+ },
70
+ {
71
+ "epoch": 0.02,
72
+ "learning_rate": 0.0002,
73
+ "loss": 1.0481,
74
+ "step": 11
75
+ },
76
+ {
77
+ "epoch": 0.02,
78
+ "learning_rate": 0.0002,
79
+ "loss": 0.7954,
80
+ "step": 12
81
+ },
82
+ {
83
+ "epoch": 0.02,
84
+ "learning_rate": 0.0002,
85
+ "loss": 1.3301,
86
+ "step": 13
87
+ },
88
+ {
89
+ "epoch": 0.02,
90
+ "learning_rate": 0.0002,
91
+ "loss": 0.9544,
92
+ "step": 14
93
+ },
94
+ {
95
+ "epoch": 0.02,
96
+ "learning_rate": 0.0002,
97
+ "loss": 0.8664,
98
+ "step": 15
99
+ },
100
+ {
101
+ "epoch": 0.02,
102
+ "learning_rate": 0.0002,
103
+ "loss": 1.1004,
104
+ "step": 16
105
+ },
106
+ {
107
+ "epoch": 0.02,
108
+ "learning_rate": 0.0002,
109
+ "loss": 0.6603,
110
+ "step": 17
111
+ },
112
+ {
113
+ "epoch": 0.03,
114
+ "learning_rate": 0.0002,
115
+ "loss": 0.5278,
116
+ "step": 18
117
+ },
118
+ {
119
+ "epoch": 0.03,
120
+ "learning_rate": 0.0002,
121
+ "loss": 0.5971,
122
+ "step": 19
123
+ },
124
+ {
125
+ "epoch": 0.03,
126
+ "learning_rate": 0.0002,
127
+ "loss": 0.6106,
128
+ "step": 20
129
+ },
130
+ {
131
+ "epoch": 0.03,
132
+ "learning_rate": 0.0002,
133
+ "loss": 0.9927,
134
+ "step": 21
135
+ },
136
+ {
137
+ "epoch": 0.03,
138
+ "learning_rate": 0.0002,
139
+ "loss": 0.7429,
140
+ "step": 22
141
+ },
142
+ {
143
+ "epoch": 0.03,
144
+ "learning_rate": 0.0002,
145
+ "loss": 0.5403,
146
+ "step": 23
147
+ },
148
+ {
149
+ "epoch": 0.03,
150
+ "learning_rate": 0.0002,
151
+ "loss": 0.8739,
152
+ "step": 24
153
+ },
154
+ {
155
+ "epoch": 0.04,
156
+ "learning_rate": 0.0002,
157
+ "loss": 0.7539,
158
+ "step": 25
159
+ },
160
+ {
161
+ "epoch": 0.04,
162
+ "learning_rate": 0.0002,
163
+ "loss": 0.7598,
164
+ "step": 26
165
+ },
166
+ {
167
+ "epoch": 0.04,
168
+ "learning_rate": 0.0002,
169
+ "loss": 0.9449,
170
+ "step": 27
171
+ },
172
+ {
173
+ "epoch": 0.04,
174
+ "learning_rate": 0.0002,
175
+ "loss": 0.7586,
176
+ "step": 28
177
+ },
178
+ {
179
+ "epoch": 0.04,
180
+ "learning_rate": 0.0002,
181
+ "loss": 0.5068,
182
+ "step": 29
183
+ },
184
+ {
185
+ "epoch": 0.04,
186
+ "learning_rate": 0.0002,
187
+ "loss": 0.6072,
188
+ "step": 30
189
+ },
190
+ {
191
+ "epoch": 0.04,
192
+ "learning_rate": 0.0002,
193
+ "loss": 0.6695,
194
+ "step": 31
195
+ },
196
+ {
197
+ "epoch": 0.04,
198
+ "learning_rate": 0.0002,
199
+ "loss": 0.9185,
200
+ "step": 32
201
+ },
202
+ {
203
+ "epoch": 0.05,
204
+ "learning_rate": 0.0002,
205
+ "loss": 0.6406,
206
+ "step": 33
207
+ },
208
+ {
209
+ "epoch": 0.05,
210
+ "learning_rate": 0.0002,
211
+ "loss": 1.2219,
212
+ "step": 34
213
+ },
214
+ {
215
+ "epoch": 0.05,
216
+ "learning_rate": 0.0002,
217
+ "loss": 0.8394,
218
+ "step": 35
219
+ },
220
+ {
221
+ "epoch": 0.05,
222
+ "learning_rate": 0.0002,
223
+ "loss": 0.6016,
224
+ "step": 36
225
+ },
226
+ {
227
+ "epoch": 0.05,
228
+ "learning_rate": 0.0002,
229
+ "loss": 0.3244,
230
+ "step": 37
231
+ },
232
+ {
233
+ "epoch": 0.05,
234
+ "learning_rate": 0.0002,
235
+ "loss": 0.5451,
236
+ "step": 38
237
+ },
238
+ {
239
+ "epoch": 0.05,
240
+ "learning_rate": 0.0002,
241
+ "loss": 0.7036,
242
+ "step": 39
243
+ },
244
+ {
245
+ "epoch": 0.06,
246
+ "learning_rate": 0.0002,
247
+ "loss": 0.6548,
248
+ "step": 40
249
+ },
250
+ {
251
+ "epoch": 0.06,
252
+ "learning_rate": 0.0002,
253
+ "loss": 0.9478,
254
+ "step": 41
255
+ },
256
+ {
257
+ "epoch": 0.06,
258
+ "learning_rate": 0.0002,
259
+ "loss": 0.9163,
260
+ "step": 42
261
+ },
262
+ {
263
+ "epoch": 0.06,
264
+ "learning_rate": 0.0002,
265
+ "loss": 0.9001,
266
+ "step": 43
267
+ },
268
+ {
269
+ "epoch": 0.06,
270
+ "learning_rate": 0.0002,
271
+ "loss": 0.7748,
272
+ "step": 44
273
+ },
274
+ {
275
+ "epoch": 0.06,
276
+ "learning_rate": 0.0002,
277
+ "loss": 0.7414,
278
+ "step": 45
279
+ },
280
+ {
281
+ "epoch": 0.06,
282
+ "learning_rate": 0.0002,
283
+ "loss": 0.4948,
284
+ "step": 46
285
+ },
286
+ {
287
+ "epoch": 0.07,
288
+ "learning_rate": 0.0002,
289
+ "loss": 0.7041,
290
+ "step": 47
291
+ },
292
+ {
293
+ "epoch": 0.07,
294
+ "learning_rate": 0.0002,
295
+ "loss": 0.6931,
296
+ "step": 48
297
+ },
298
+ {
299
+ "epoch": 0.07,
300
+ "learning_rate": 0.0002,
301
+ "loss": 0.7196,
302
+ "step": 49
303
+ },
304
+ {
305
+ "epoch": 0.07,
306
+ "learning_rate": 0.0002,
307
+ "loss": 0.6884,
308
+ "step": 50
309
+ },
310
+ {
311
+ "epoch": 0.07,
312
+ "learning_rate": 0.0002,
313
+ "loss": 0.5539,
314
+ "step": 51
315
+ },
316
+ {
317
+ "epoch": 0.07,
318
+ "learning_rate": 0.0002,
319
+ "loss": 0.6652,
320
+ "step": 52
321
+ },
322
+ {
323
+ "epoch": 0.07,
324
+ "learning_rate": 0.0002,
325
+ "loss": 0.9422,
326
+ "step": 53
327
+ },
328
+ {
329
+ "epoch": 0.08,
330
+ "learning_rate": 0.0002,
331
+ "loss": 0.447,
332
+ "step": 54
333
+ },
334
+ {
335
+ "epoch": 0.08,
336
+ "learning_rate": 0.0002,
337
+ "loss": 0.7642,
338
+ "step": 55
339
+ },
340
+ {
341
+ "epoch": 0.08,
342
+ "learning_rate": 0.0002,
343
+ "loss": 0.4434,
344
+ "step": 56
345
+ },
346
+ {
347
+ "epoch": 0.08,
348
+ "learning_rate": 0.0002,
349
+ "loss": 0.7617,
350
+ "step": 57
351
+ },
352
+ {
353
+ "epoch": 0.08,
354
+ "learning_rate": 0.0002,
355
+ "loss": 0.5312,
356
+ "step": 58
357
+ },
358
+ {
359
+ "epoch": 0.08,
360
+ "learning_rate": 0.0002,
361
+ "loss": 0.7388,
362
+ "step": 59
363
+ },
364
+ {
365
+ "epoch": 0.08,
366
+ "learning_rate": 0.0002,
367
+ "loss": 0.9614,
368
+ "step": 60
369
+ },
370
+ {
371
+ "epoch": 0.09,
372
+ "learning_rate": 0.0002,
373
+ "loss": 1.2599,
374
+ "step": 61
375
+ },
376
+ {
377
+ "epoch": 0.09,
378
+ "learning_rate": 0.0002,
379
+ "loss": 0.595,
380
+ "step": 62
381
+ },
382
+ {
383
+ "epoch": 0.09,
384
+ "learning_rate": 0.0002,
385
+ "loss": 0.7218,
386
+ "step": 63
387
+ },
388
+ {
389
+ "epoch": 0.09,
390
+ "learning_rate": 0.0002,
391
+ "loss": 0.5194,
392
+ "step": 64
393
+ },
394
+ {
395
+ "epoch": 0.09,
396
+ "learning_rate": 0.0002,
397
+ "loss": 0.5519,
398
+ "step": 65
399
+ },
400
+ {
401
+ "epoch": 0.09,
402
+ "learning_rate": 0.0002,
403
+ "loss": 0.5712,
404
+ "step": 66
405
+ },
406
+ {
407
+ "epoch": 0.09,
408
+ "learning_rate": 0.0002,
409
+ "loss": 0.6011,
410
+ "step": 67
411
+ },
412
+ {
413
+ "epoch": 0.1,
414
+ "learning_rate": 0.0002,
415
+ "loss": 0.6643,
416
+ "step": 68
417
+ },
418
+ {
419
+ "epoch": 0.1,
420
+ "learning_rate": 0.0002,
421
+ "loss": 0.7581,
422
+ "step": 69
423
+ },
424
+ {
425
+ "epoch": 0.1,
426
+ "learning_rate": 0.0002,
427
+ "loss": 1.2109,
428
+ "step": 70
429
+ },
430
+ {
431
+ "epoch": 0.1,
432
+ "learning_rate": 0.0002,
433
+ "loss": 0.8814,
434
+ "step": 71
435
+ },
436
+ {
437
+ "epoch": 0.1,
438
+ "learning_rate": 0.0002,
439
+ "loss": 0.507,
440
+ "step": 72
441
+ },
442
+ {
443
+ "epoch": 0.1,
444
+ "learning_rate": 0.0002,
445
+ "loss": 0.7569,
446
+ "step": 73
447
+ },
448
+ {
449
+ "epoch": 0.1,
450
+ "learning_rate": 0.0002,
451
+ "loss": 0.7431,
452
+ "step": 74
453
+ },
454
+ {
455
+ "epoch": 0.11,
456
+ "learning_rate": 0.0002,
457
+ "loss": 0.7837,
458
+ "step": 75
459
+ },
460
+ {
461
+ "epoch": 0.11,
462
+ "learning_rate": 0.0002,
463
+ "loss": 0.5242,
464
+ "step": 76
465
+ },
466
+ {
467
+ "epoch": 0.11,
468
+ "learning_rate": 0.0002,
469
+ "loss": 0.3062,
470
+ "step": 77
471
+ },
472
+ {
473
+ "epoch": 0.11,
474
+ "learning_rate": 0.0002,
475
+ "loss": 0.5109,
476
+ "step": 78
477
+ },
478
+ {
479
+ "epoch": 0.11,
480
+ "learning_rate": 0.0002,
481
+ "loss": 0.8998,
482
+ "step": 79
483
+ },
484
+ {
485
+ "epoch": 0.11,
486
+ "learning_rate": 0.0002,
487
+ "loss": 0.875,
488
+ "step": 80
489
+ },
490
+ {
491
+ "epoch": 0.11,
492
+ "learning_rate": 0.0002,
493
+ "loss": 0.7363,
494
+ "step": 81
495
+ },
496
+ {
497
+ "epoch": 0.12,
498
+ "learning_rate": 0.0002,
499
+ "loss": 0.6465,
500
+ "step": 82
501
+ },
502
+ {
503
+ "epoch": 0.12,
504
+ "learning_rate": 0.0002,
505
+ "loss": 0.8307,
506
+ "step": 83
507
+ },
508
+ {
509
+ "epoch": 0.12,
510
+ "learning_rate": 0.0002,
511
+ "loss": 0.6284,
512
+ "step": 84
513
+ },
514
+ {
515
+ "epoch": 0.12,
516
+ "learning_rate": 0.0002,
517
+ "loss": 0.5862,
518
+ "step": 85
519
+ },
520
+ {
521
+ "epoch": 0.12,
522
+ "learning_rate": 0.0002,
523
+ "loss": 0.7502,
524
+ "step": 86
525
+ },
526
+ {
527
+ "epoch": 0.12,
528
+ "learning_rate": 0.0002,
529
+ "loss": 0.6251,
530
+ "step": 87
531
+ },
532
+ {
533
+ "epoch": 0.12,
534
+ "learning_rate": 0.0002,
535
+ "loss": 0.7613,
536
+ "step": 88
537
+ },
538
+ {
539
+ "epoch": 0.12,
540
+ "learning_rate": 0.0002,
541
+ "loss": 0.6527,
542
+ "step": 89
543
+ },
544
+ {
545
+ "epoch": 0.13,
546
+ "learning_rate": 0.0002,
547
+ "loss": 0.5793,
548
+ "step": 90
549
+ },
550
+ {
551
+ "epoch": 0.13,
552
+ "learning_rate": 0.0002,
553
+ "loss": 0.459,
554
+ "step": 91
555
+ },
556
+ {
557
+ "epoch": 0.13,
558
+ "learning_rate": 0.0002,
559
+ "loss": 0.7156,
560
+ "step": 92
561
+ },
562
+ {
563
+ "epoch": 0.13,
564
+ "learning_rate": 0.0002,
565
+ "loss": 0.6078,
566
+ "step": 93
567
+ },
568
+ {
569
+ "epoch": 0.13,
570
+ "learning_rate": 0.0002,
571
+ "loss": 0.2572,
572
+ "step": 94
573
+ },
574
+ {
575
+ "epoch": 0.13,
576
+ "learning_rate": 0.0002,
577
+ "loss": 0.5447,
578
+ "step": 95
579
+ },
580
+ {
581
+ "epoch": 0.13,
582
+ "learning_rate": 0.0002,
583
+ "loss": 0.6109,
584
+ "step": 96
585
+ },
586
+ {
587
+ "epoch": 0.14,
588
+ "learning_rate": 0.0002,
589
+ "loss": 1.0691,
590
+ "step": 97
591
+ },
592
+ {
593
+ "epoch": 0.14,
594
+ "learning_rate": 0.0002,
595
+ "loss": 0.3149,
596
+ "step": 98
597
+ },
598
+ {
599
+ "epoch": 0.14,
600
+ "learning_rate": 0.0002,
601
+ "loss": 0.5433,
602
+ "step": 99
603
+ },
604
+ {
605
+ "epoch": 0.14,
606
+ "learning_rate": 0.0002,
607
+ "loss": 0.7371,
608
+ "step": 100
609
+ },
610
+ {
611
+ "epoch": 0.14,
612
+ "learning_rate": 0.0002,
613
+ "loss": 0.8157,
614
+ "step": 101
615
+ },
616
+ {
617
+ "epoch": 0.14,
618
+ "learning_rate": 0.0002,
619
+ "loss": 0.5899,
620
+ "step": 102
621
+ },
622
+ {
623
+ "epoch": 0.14,
624
+ "learning_rate": 0.0002,
625
+ "loss": 0.6743,
626
+ "step": 103
627
+ },
628
+ {
629
+ "epoch": 0.15,
630
+ "learning_rate": 0.0002,
631
+ "loss": 0.9741,
632
+ "step": 104
633
+ },
634
+ {
635
+ "epoch": 0.15,
636
+ "learning_rate": 0.0002,
637
+ "loss": 0.5697,
638
+ "step": 105
639
+ },
640
+ {
641
+ "epoch": 0.15,
642
+ "learning_rate": 0.0002,
643
+ "loss": 0.6596,
644
+ "step": 106
645
+ },
646
+ {
647
+ "epoch": 0.15,
648
+ "learning_rate": 0.0002,
649
+ "loss": 0.7502,
650
+ "step": 107
651
+ },
652
+ {
653
+ "epoch": 0.15,
654
+ "learning_rate": 0.0002,
655
+ "loss": 0.6138,
656
+ "step": 108
657
+ },
658
+ {
659
+ "epoch": 0.15,
660
+ "learning_rate": 0.0002,
661
+ "loss": 0.6065,
662
+ "step": 109
663
+ },
664
+ {
665
+ "epoch": 0.15,
666
+ "learning_rate": 0.0002,
667
+ "loss": 0.6661,
668
+ "step": 110
669
+ },
670
+ {
671
+ "epoch": 0.16,
672
+ "learning_rate": 0.0002,
673
+ "loss": 0.987,
674
+ "step": 111
675
+ },
676
+ {
677
+ "epoch": 0.16,
678
+ "learning_rate": 0.0002,
679
+ "loss": 0.773,
680
+ "step": 112
681
+ },
682
+ {
683
+ "epoch": 0.16,
684
+ "learning_rate": 0.0002,
685
+ "loss": 0.4769,
686
+ "step": 113
687
+ },
688
+ {
689
+ "epoch": 0.16,
690
+ "learning_rate": 0.0002,
691
+ "loss": 0.8634,
692
+ "step": 114
693
+ },
694
+ {
695
+ "epoch": 0.16,
696
+ "learning_rate": 0.0002,
697
+ "loss": 0.9495,
698
+ "step": 115
699
+ },
700
+ {
701
+ "epoch": 0.16,
702
+ "learning_rate": 0.0002,
703
+ "loss": 0.6708,
704
+ "step": 116
705
+ },
706
+ {
707
+ "epoch": 0.16,
708
+ "learning_rate": 0.0002,
709
+ "loss": 0.7206,
710
+ "step": 117
711
+ },
712
+ {
713
+ "epoch": 0.17,
714
+ "learning_rate": 0.0002,
715
+ "loss": 0.628,
716
+ "step": 118
717
+ },
718
+ {
719
+ "epoch": 0.17,
720
+ "learning_rate": 0.0002,
721
+ "loss": 0.7703,
722
+ "step": 119
723
+ },
724
+ {
725
+ "epoch": 0.17,
726
+ "learning_rate": 0.0002,
727
+ "loss": 0.6506,
728
+ "step": 120
729
+ },
730
+ {
731
+ "epoch": 0.17,
732
+ "learning_rate": 0.0002,
733
+ "loss": 0.8484,
734
+ "step": 121
735
+ },
736
+ {
737
+ "epoch": 0.17,
738
+ "learning_rate": 0.0002,
739
+ "loss": 0.8516,
740
+ "step": 122
741
+ },
742
+ {
743
+ "epoch": 0.17,
744
+ "learning_rate": 0.0002,
745
+ "loss": 0.4612,
746
+ "step": 123
747
+ },
748
+ {
749
+ "epoch": 0.17,
750
+ "learning_rate": 0.0002,
751
+ "loss": 0.5995,
752
+ "step": 124
753
+ },
754
+ {
755
+ "epoch": 0.18,
756
+ "learning_rate": 0.0002,
757
+ "loss": 0.6679,
758
+ "step": 125
759
+ },
760
+ {
761
+ "epoch": 0.18,
762
+ "learning_rate": 0.0002,
763
+ "loss": 0.8091,
764
+ "step": 126
765
+ },
766
+ {
767
+ "epoch": 0.18,
768
+ "learning_rate": 0.0002,
769
+ "loss": 0.6702,
770
+ "step": 127
771
+ },
772
+ {
773
+ "epoch": 0.18,
774
+ "learning_rate": 0.0002,
775
+ "loss": 0.7588,
776
+ "step": 128
777
+ },
778
+ {
779
+ "epoch": 0.18,
780
+ "learning_rate": 0.0002,
781
+ "loss": 0.5301,
782
+ "step": 129
783
+ },
784
+ {
785
+ "epoch": 0.18,
786
+ "learning_rate": 0.0002,
787
+ "loss": 0.4867,
788
+ "step": 130
789
+ },
790
+ {
791
+ "epoch": 0.18,
792
+ "learning_rate": 0.0002,
793
+ "loss": 1.0696,
794
+ "step": 131
795
+ },
796
+ {
797
+ "epoch": 0.19,
798
+ "learning_rate": 0.0002,
799
+ "loss": 0.3769,
800
+ "step": 132
801
+ },
802
+ {
803
+ "epoch": 0.19,
804
+ "learning_rate": 0.0002,
805
+ "loss": 0.558,
806
+ "step": 133
807
+ },
808
+ {
809
+ "epoch": 0.19,
810
+ "learning_rate": 0.0002,
811
+ "loss": 0.7511,
812
+ "step": 134
813
+ },
814
+ {
815
+ "epoch": 0.19,
816
+ "learning_rate": 0.0002,
817
+ "loss": 0.3991,
818
+ "step": 135
819
+ },
820
+ {
821
+ "epoch": 0.19,
822
+ "learning_rate": 0.0002,
823
+ "loss": 0.6713,
824
+ "step": 136
825
+ },
826
+ {
827
+ "epoch": 0.19,
828
+ "learning_rate": 0.0002,
829
+ "loss": 0.4538,
830
+ "step": 137
831
+ },
832
+ {
833
+ "epoch": 0.19,
834
+ "learning_rate": 0.0002,
835
+ "loss": 0.5503,
836
+ "step": 138
837
+ },
838
+ {
839
+ "epoch": 0.2,
840
+ "learning_rate": 0.0002,
841
+ "loss": 0.6534,
842
+ "step": 139
843
+ },
844
+ {
845
+ "epoch": 0.2,
846
+ "learning_rate": 0.0002,
847
+ "loss": 0.4622,
848
+ "step": 140
849
+ },
850
+ {
851
+ "epoch": 0.2,
852
+ "learning_rate": 0.0002,
853
+ "loss": 0.5731,
854
+ "step": 141
855
+ },
856
+ {
857
+ "epoch": 0.2,
858
+ "learning_rate": 0.0002,
859
+ "loss": 0.7238,
860
+ "step": 142
861
+ },
862
+ {
863
+ "epoch": 0.2,
864
+ "learning_rate": 0.0002,
865
+ "loss": 0.8143,
866
+ "step": 143
867
+ },
868
+ {
869
+ "epoch": 0.2,
870
+ "learning_rate": 0.0002,
871
+ "loss": 0.539,
872
+ "step": 144
873
+ },
874
+ {
875
+ "epoch": 0.2,
876
+ "learning_rate": 0.0002,
877
+ "loss": 0.447,
878
+ "step": 145
879
+ },
880
+ {
881
+ "epoch": 0.2,
882
+ "learning_rate": 0.0002,
883
+ "loss": 0.4711,
884
+ "step": 146
885
+ },
886
+ {
887
+ "epoch": 0.21,
888
+ "learning_rate": 0.0002,
889
+ "loss": 0.5584,
890
+ "step": 147
891
+ },
892
+ {
893
+ "epoch": 0.21,
894
+ "learning_rate": 0.0002,
895
+ "loss": 0.5887,
896
+ "step": 148
897
+ },
898
+ {
899
+ "epoch": 0.21,
900
+ "learning_rate": 0.0002,
901
+ "loss": 0.3808,
902
+ "step": 149
903
+ },
904
+ {
905
+ "epoch": 0.21,
906
+ "learning_rate": 0.0002,
907
+ "loss": 0.5771,
908
+ "step": 150
909
+ },
910
+ {
911
+ "epoch": 0.21,
912
+ "learning_rate": 0.0002,
913
+ "loss": 0.5276,
914
+ "step": 151
915
+ },
916
+ {
917
+ "epoch": 0.21,
918
+ "learning_rate": 0.0002,
919
+ "loss": 0.4132,
920
+ "step": 152
921
+ },
922
+ {
923
+ "epoch": 0.21,
924
+ "learning_rate": 0.0002,
925
+ "loss": 0.9056,
926
+ "step": 153
927
+ },
928
+ {
929
+ "epoch": 0.22,
930
+ "learning_rate": 0.0002,
931
+ "loss": 0.5737,
932
+ "step": 154
933
+ },
934
+ {
935
+ "epoch": 0.22,
936
+ "learning_rate": 0.0002,
937
+ "loss": 0.6992,
938
+ "step": 155
939
+ },
940
+ {
941
+ "epoch": 0.22,
942
+ "learning_rate": 0.0002,
943
+ "loss": 0.4023,
944
+ "step": 156
945
+ },
946
+ {
947
+ "epoch": 0.22,
948
+ "learning_rate": 0.0002,
949
+ "loss": 0.6238,
950
+ "step": 157
951
+ },
952
+ {
953
+ "epoch": 0.22,
954
+ "learning_rate": 0.0002,
955
+ "loss": 0.5521,
956
+ "step": 158
957
+ },
958
+ {
959
+ "epoch": 0.22,
960
+ "learning_rate": 0.0002,
961
+ "loss": 0.6641,
962
+ "step": 159
963
+ },
964
+ {
965
+ "epoch": 0.22,
966
+ "learning_rate": 0.0002,
967
+ "loss": 0.4478,
968
+ "step": 160
969
+ },
970
+ {
971
+ "epoch": 0.23,
972
+ "learning_rate": 0.0002,
973
+ "loss": 0.865,
974
+ "step": 161
975
+ },
976
+ {
977
+ "epoch": 0.23,
978
+ "learning_rate": 0.0002,
979
+ "loss": 0.4332,
980
+ "step": 162
981
+ },
982
+ {
983
+ "epoch": 0.23,
984
+ "learning_rate": 0.0002,
985
+ "loss": 0.8205,
986
+ "step": 163
987
+ },
988
+ {
989
+ "epoch": 0.23,
990
+ "learning_rate": 0.0002,
991
+ "loss": 0.6539,
992
+ "step": 164
993
+ },
994
+ {
995
+ "epoch": 0.23,
996
+ "learning_rate": 0.0002,
997
+ "loss": 0.6957,
998
+ "step": 165
999
+ },
1000
+ {
1001
+ "epoch": 0.23,
1002
+ "learning_rate": 0.0002,
1003
+ "loss": 0.6279,
1004
+ "step": 166
1005
+ },
1006
+ {
1007
+ "epoch": 0.23,
1008
+ "learning_rate": 0.0002,
1009
+ "loss": 0.6073,
1010
+ "step": 167
1011
+ },
1012
+ {
1013
+ "epoch": 0.24,
1014
+ "learning_rate": 0.0002,
1015
+ "loss": 0.5831,
1016
+ "step": 168
1017
+ },
1018
+ {
1019
+ "epoch": 0.24,
1020
+ "learning_rate": 0.0002,
1021
+ "loss": 0.5383,
1022
+ "step": 169
1023
+ },
1024
+ {
1025
+ "epoch": 0.24,
1026
+ "learning_rate": 0.0002,
1027
+ "loss": 1.0065,
1028
+ "step": 170
1029
+ },
1030
+ {
1031
+ "epoch": 0.24,
1032
+ "learning_rate": 0.0002,
1033
+ "loss": 0.5239,
1034
+ "step": 171
1035
+ },
1036
+ {
1037
+ "epoch": 0.24,
1038
+ "learning_rate": 0.0002,
1039
+ "loss": 0.4246,
1040
+ "step": 172
1041
+ },
1042
+ {
1043
+ "epoch": 0.24,
1044
+ "learning_rate": 0.0002,
1045
+ "loss": 0.633,
1046
+ "step": 173
1047
+ },
1048
+ {
1049
+ "epoch": 0.24,
1050
+ "learning_rate": 0.0002,
1051
+ "loss": 0.4318,
1052
+ "step": 174
1053
+ },
1054
+ {
1055
+ "epoch": 0.25,
1056
+ "learning_rate": 0.0002,
1057
+ "loss": 0.5228,
1058
+ "step": 175
1059
+ },
1060
+ {
1061
+ "epoch": 0.25,
1062
+ "learning_rate": 0.0002,
1063
+ "loss": 0.791,
1064
+ "step": 176
1065
+ },
1066
+ {
1067
+ "epoch": 0.25,
1068
+ "learning_rate": 0.0002,
1069
+ "loss": 0.6141,
1070
+ "step": 177
1071
+ },
1072
+ {
1073
+ "epoch": 0.25,
1074
+ "learning_rate": 0.0002,
1075
+ "loss": 0.4894,
1076
+ "step": 178
1077
+ },
1078
+ {
1079
+ "epoch": 0.25,
1080
+ "learning_rate": 0.0002,
1081
+ "loss": 1.0344,
1082
+ "step": 179
1083
+ },
1084
+ {
1085
+ "epoch": 0.25,
1086
+ "learning_rate": 0.0002,
1087
+ "loss": 0.7332,
1088
+ "step": 180
1089
+ },
1090
+ {
1091
+ "epoch": 0.25,
1092
+ "learning_rate": 0.0002,
1093
+ "loss": 0.6845,
1094
+ "step": 181
1095
+ },
1096
+ {
1097
+ "epoch": 0.26,
1098
+ "learning_rate": 0.0002,
1099
+ "loss": 0.8027,
1100
+ "step": 182
1101
+ },
1102
+ {
1103
+ "epoch": 0.26,
1104
+ "learning_rate": 0.0002,
1105
+ "loss": 0.6537,
1106
+ "step": 183
1107
+ },
1108
+ {
1109
+ "epoch": 0.26,
1110
+ "learning_rate": 0.0002,
1111
+ "loss": 0.7333,
1112
+ "step": 184
1113
+ },
1114
+ {
1115
+ "epoch": 0.26,
1116
+ "learning_rate": 0.0002,
1117
+ "loss": 0.6149,
1118
+ "step": 185
1119
+ },
1120
+ {
1121
+ "epoch": 0.26,
1122
+ "learning_rate": 0.0002,
1123
+ "loss": 1.0586,
1124
+ "step": 186
1125
+ },
1126
+ {
1127
+ "epoch": 0.26,
1128
+ "learning_rate": 0.0002,
1129
+ "loss": 0.6911,
1130
+ "step": 187
1131
+ },
1132
+ {
1133
+ "epoch": 0.26,
1134
+ "learning_rate": 0.0002,
1135
+ "loss": 0.75,
1136
+ "step": 188
1137
+ },
1138
+ {
1139
+ "epoch": 0.27,
1140
+ "learning_rate": 0.0002,
1141
+ "loss": 0.4598,
1142
+ "step": 189
1143
+ },
1144
+ {
1145
+ "epoch": 0.27,
1146
+ "learning_rate": 0.0002,
1147
+ "loss": 0.5292,
1148
+ "step": 190
1149
+ },
1150
+ {
1151
+ "epoch": 0.27,
1152
+ "learning_rate": 0.0002,
1153
+ "loss": 0.6164,
1154
+ "step": 191
1155
+ },
1156
+ {
1157
+ "epoch": 0.27,
1158
+ "learning_rate": 0.0002,
1159
+ "loss": 0.3216,
1160
+ "step": 192
1161
+ },
1162
+ {
1163
+ "epoch": 0.27,
1164
+ "learning_rate": 0.0002,
1165
+ "loss": 0.6229,
1166
+ "step": 193
1167
+ },
1168
+ {
1169
+ "epoch": 0.27,
1170
+ "learning_rate": 0.0002,
1171
+ "loss": 0.6054,
1172
+ "step": 194
1173
+ },
1174
+ {
1175
+ "epoch": 0.27,
1176
+ "learning_rate": 0.0002,
1177
+ "loss": 0.4821,
1178
+ "step": 195
1179
+ },
1180
+ {
1181
+ "epoch": 0.28,
1182
+ "learning_rate": 0.0002,
1183
+ "loss": 0.8481,
1184
+ "step": 196
1185
+ },
1186
+ {
1187
+ "epoch": 0.28,
1188
+ "learning_rate": 0.0002,
1189
+ "loss": 0.633,
1190
+ "step": 197
1191
+ },
1192
+ {
1193
+ "epoch": 0.28,
1194
+ "learning_rate": 0.0002,
1195
+ "loss": 0.4537,
1196
+ "step": 198
1197
+ },
1198
+ {
1199
+ "epoch": 0.28,
1200
+ "learning_rate": 0.0002,
1201
+ "loss": 0.6689,
1202
+ "step": 199
1203
+ },
1204
+ {
1205
+ "epoch": 0.28,
1206
+ "learning_rate": 0.0002,
1207
+ "loss": 0.8891,
1208
+ "step": 200
1209
+ },
1210
+ {
1211
+ "epoch": 0.28,
1212
+ "eval_loss": 0.6127411127090454,
1213
+ "eval_runtime": 397.1016,
1214
+ "eval_samples_per_second": 2.518,
1215
+ "eval_steps_per_second": 1.259,
1216
+ "step": 200
1217
+ },
1218
+ {
1219
+ "epoch": 0.28,
1220
+ "mmlu_eval_accuracy": 0.49968036766977914,
1221
+ "mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
1222
+ "mmlu_eval_accuracy_anatomy": 0.5,
1223
+ "mmlu_eval_accuracy_astronomy": 0.375,
1224
+ "mmlu_eval_accuracy_business_ethics": 0.2727272727272727,
1225
+ "mmlu_eval_accuracy_clinical_knowledge": 0.4482758620689655,
1226
+ "mmlu_eval_accuracy_college_biology": 0.5,
1227
+ "mmlu_eval_accuracy_college_chemistry": 0.25,
1228
+ "mmlu_eval_accuracy_college_computer_science": 0.5454545454545454,
1229
+ "mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
1230
+ "mmlu_eval_accuracy_college_medicine": 0.5454545454545454,
1231
+ "mmlu_eval_accuracy_college_physics": 0.5454545454545454,
1232
+ "mmlu_eval_accuracy_computer_security": 0.45454545454545453,
1233
+ "mmlu_eval_accuracy_conceptual_physics": 0.2692307692307692,
1234
+ "mmlu_eval_accuracy_econometrics": 0.3333333333333333,
1235
+ "mmlu_eval_accuracy_electrical_engineering": 0.25,
1236
+ "mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
1237
+ "mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
1238
+ "mmlu_eval_accuracy_global_facts": 0.3,
1239
+ "mmlu_eval_accuracy_high_school_biology": 0.46875,
1240
+ "mmlu_eval_accuracy_high_school_chemistry": 0.2727272727272727,
1241
+ "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
1242
+ "mmlu_eval_accuracy_high_school_european_history": 0.6666666666666666,
1243
+ "mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
1244
+ "mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
1245
+ "mmlu_eval_accuracy_high_school_macroeconomics": 0.3488372093023256,
1246
+ "mmlu_eval_accuracy_high_school_mathematics": 0.1724137931034483,
1247
+ "mmlu_eval_accuracy_high_school_microeconomics": 0.5384615384615384,
1248
+ "mmlu_eval_accuracy_high_school_physics": 0.35294117647058826,
1249
+ "mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
1250
+ "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
1251
+ "mmlu_eval_accuracy_high_school_us_history": 0.8181818181818182,
1252
+ "mmlu_eval_accuracy_high_school_world_history": 0.6153846153846154,
1253
+ "mmlu_eval_accuracy_human_aging": 0.4782608695652174,
1254
+ "mmlu_eval_accuracy_human_sexuality": 0.5,
1255
+ "mmlu_eval_accuracy_international_law": 0.8461538461538461,
1256
+ "mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
1257
+ "mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
1258
+ "mmlu_eval_accuracy_machine_learning": 0.6363636363636364,
1259
+ "mmlu_eval_accuracy_management": 0.7272727272727273,
1260
+ "mmlu_eval_accuracy_marketing": 0.68,
1261
+ "mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
1262
+ "mmlu_eval_accuracy_miscellaneous": 0.6511627906976745,
1263
+ "mmlu_eval_accuracy_moral_disputes": 0.5,
1264
+ "mmlu_eval_accuracy_moral_scenarios": 0.21,
1265
+ "mmlu_eval_accuracy_nutrition": 0.696969696969697,
1266
+ "mmlu_eval_accuracy_philosophy": 0.5882352941176471,
1267
+ "mmlu_eval_accuracy_prehistory": 0.5428571428571428,
1268
+ "mmlu_eval_accuracy_professional_accounting": 0.3225806451612903,
1269
+ "mmlu_eval_accuracy_professional_law": 0.37058823529411766,
1270
+ "mmlu_eval_accuracy_professional_medicine": 0.4838709677419355,
1271
+ "mmlu_eval_accuracy_professional_psychology": 0.4492753623188406,
1272
+ "mmlu_eval_accuracy_public_relations": 0.4166666666666667,
1273
+ "mmlu_eval_accuracy_security_studies": 0.3333333333333333,
1274
+ "mmlu_eval_accuracy_sociology": 0.7272727272727273,
1275
+ "mmlu_eval_accuracy_us_foreign_policy": 0.5454545454545454,
1276
+ "mmlu_eval_accuracy_virology": 0.4444444444444444,
1277
+ "mmlu_eval_accuracy_world_religions": 0.7368421052631579,
1278
+ "mmlu_loss": 0.9665496518188607,
1279
+ "step": 200
1280
+ },
1281
+ {
1282
+ "epoch": 0.28,
1283
+ "learning_rate": 0.0002,
1284
+ "loss": 0.8622,
1285
+ "step": 201
1286
+ },
1287
+ {
1288
+ "epoch": 0.28,
1289
+ "learning_rate": 0.0002,
1290
+ "loss": 0.7277,
1291
+ "step": 202
1292
+ },
1293
+ {
1294
+ "epoch": 0.29,
1295
+ "learning_rate": 0.0002,
1296
+ "loss": 0.7358,
1297
+ "step": 203
1298
+ },
1299
+ {
1300
+ "epoch": 0.29,
1301
+ "learning_rate": 0.0002,
1302
+ "loss": 0.6171,
1303
+ "step": 204
1304
+ },
1305
+ {
1306
+ "epoch": 0.29,
1307
+ "learning_rate": 0.0002,
1308
+ "loss": 0.7573,
1309
+ "step": 205
1310
+ },
1311
+ {
1312
+ "epoch": 0.29,
1313
+ "learning_rate": 0.0002,
1314
+ "loss": 0.7947,
1315
+ "step": 206
1316
+ },
1317
+ {
1318
+ "epoch": 0.29,
1319
+ "learning_rate": 0.0002,
1320
+ "loss": 0.438,
1321
+ "step": 207
1322
+ },
1323
+ {
1324
+ "epoch": 0.29,
1325
+ "learning_rate": 0.0002,
1326
+ "loss": 0.4314,
1327
+ "step": 208
1328
+ },
1329
+ {
1330
+ "epoch": 0.29,
1331
+ "learning_rate": 0.0002,
1332
+ "loss": 0.7751,
1333
+ "step": 209
1334
+ },
1335
+ {
1336
+ "epoch": 0.29,
1337
+ "learning_rate": 0.0002,
1338
+ "loss": 0.7193,
1339
+ "step": 210
1340
+ },
1341
+ {
1342
+ "epoch": 0.3,
1343
+ "learning_rate": 0.0002,
1344
+ "loss": 0.5671,
1345
+ "step": 211
1346
+ },
1347
+ {
1348
+ "epoch": 0.3,
1349
+ "learning_rate": 0.0002,
1350
+ "loss": 0.6712,
1351
+ "step": 212
1352
+ },
1353
+ {
1354
+ "epoch": 0.3,
1355
+ "learning_rate": 0.0002,
1356
+ "loss": 0.5761,
1357
+ "step": 213
1358
+ },
1359
+ {
1360
+ "epoch": 0.3,
1361
+ "learning_rate": 0.0002,
1362
+ "loss": 0.6589,
1363
+ "step": 214
1364
+ },
1365
+ {
1366
+ "epoch": 0.3,
1367
+ "learning_rate": 0.0002,
1368
+ "loss": 0.4178,
1369
+ "step": 215
1370
+ },
1371
+ {
1372
+ "epoch": 0.3,
1373
+ "learning_rate": 0.0002,
1374
+ "loss": 0.4622,
1375
+ "step": 216
1376
+ },
1377
+ {
1378
+ "epoch": 0.3,
1379
+ "learning_rate": 0.0002,
1380
+ "loss": 0.5189,
1381
+ "step": 217
1382
+ },
1383
+ {
1384
+ "epoch": 0.31,
1385
+ "learning_rate": 0.0002,
1386
+ "loss": 0.8789,
1387
+ "step": 218
1388
+ },
1389
+ {
1390
+ "epoch": 0.31,
1391
+ "learning_rate": 0.0002,
1392
+ "loss": 0.2529,
1393
+ "step": 219
1394
+ },
1395
+ {
1396
+ "epoch": 0.31,
1397
+ "learning_rate": 0.0002,
1398
+ "loss": 0.5861,
1399
+ "step": 220
1400
+ },
1401
+ {
1402
+ "epoch": 0.31,
1403
+ "learning_rate": 0.0002,
1404
+ "loss": 0.6578,
1405
+ "step": 221
1406
+ },
1407
+ {
1408
+ "epoch": 0.31,
1409
+ "learning_rate": 0.0002,
1410
+ "loss": 0.5289,
1411
+ "step": 222
1412
+ },
1413
+ {
1414
+ "epoch": 0.31,
1415
+ "learning_rate": 0.0002,
1416
+ "loss": 0.5529,
1417
+ "step": 223
1418
+ },
1419
+ {
1420
+ "epoch": 0.31,
1421
+ "learning_rate": 0.0002,
1422
+ "loss": 0.6496,
1423
+ "step": 224
1424
+ },
1425
+ {
1426
+ "epoch": 0.32,
1427
+ "learning_rate": 0.0002,
1428
+ "loss": 0.7043,
1429
+ "step": 225
1430
+ },
1431
+ {
1432
+ "epoch": 0.32,
1433
+ "learning_rate": 0.0002,
1434
+ "loss": 0.5955,
1435
+ "step": 226
1436
+ },
1437
+ {
1438
+ "epoch": 0.32,
1439
+ "learning_rate": 0.0002,
1440
+ "loss": 0.6942,
1441
+ "step": 227
1442
+ },
1443
+ {
1444
+ "epoch": 0.32,
1445
+ "learning_rate": 0.0002,
1446
+ "loss": 0.3691,
1447
+ "step": 228
1448
+ },
1449
+ {
1450
+ "epoch": 0.32,
1451
+ "learning_rate": 0.0002,
1452
+ "loss": 0.6348,
1453
+ "step": 229
1454
+ },
1455
+ {
1456
+ "epoch": 0.32,
1457
+ "learning_rate": 0.0002,
1458
+ "loss": 0.5973,
1459
+ "step": 230
1460
+ },
1461
+ {
1462
+ "epoch": 0.32,
1463
+ "learning_rate": 0.0002,
1464
+ "loss": 0.2706,
1465
+ "step": 231
1466
+ },
1467
+ {
1468
+ "epoch": 0.33,
1469
+ "learning_rate": 0.0002,
1470
+ "loss": 0.6844,
1471
+ "step": 232
1472
+ },
1473
+ {
1474
+ "epoch": 0.33,
1475
+ "learning_rate": 0.0002,
1476
+ "loss": 1.2182,
1477
+ "step": 233
1478
+ },
1479
+ {
1480
+ "epoch": 0.33,
1481
+ "learning_rate": 0.0002,
1482
+ "loss": 0.6845,
1483
+ "step": 234
1484
+ },
1485
+ {
1486
+ "epoch": 0.33,
1487
+ "learning_rate": 0.0002,
1488
+ "loss": 0.6271,
1489
+ "step": 235
1490
+ },
1491
+ {
1492
+ "epoch": 0.33,
1493
+ "learning_rate": 0.0002,
1494
+ "loss": 0.7151,
1495
+ "step": 236
1496
+ },
1497
+ {
1498
+ "epoch": 0.33,
1499
+ "learning_rate": 0.0002,
1500
+ "loss": 0.5568,
1501
+ "step": 237
1502
+ },
1503
+ {
1504
+ "epoch": 0.33,
1505
+ "learning_rate": 0.0002,
1506
+ "loss": 0.9098,
1507
+ "step": 238
1508
+ },
1509
+ {
1510
+ "epoch": 0.34,
1511
+ "learning_rate": 0.0002,
1512
+ "loss": 1.0819,
1513
+ "step": 239
1514
+ },
1515
+ {
1516
+ "epoch": 0.34,
1517
+ "learning_rate": 0.0002,
1518
+ "loss": 0.5894,
1519
+ "step": 240
1520
+ },
1521
+ {
1522
+ "epoch": 0.34,
1523
+ "learning_rate": 0.0002,
1524
+ "loss": 0.833,
1525
+ "step": 241
1526
+ },
1527
+ {
1528
+ "epoch": 0.34,
1529
+ "learning_rate": 0.0002,
1530
+ "loss": 0.4099,
1531
+ "step": 242
1532
+ },
1533
+ {
1534
+ "epoch": 0.34,
1535
+ "learning_rate": 0.0002,
1536
+ "loss": 0.8302,
1537
+ "step": 243
1538
+ },
1539
+ {
1540
+ "epoch": 0.34,
1541
+ "learning_rate": 0.0002,
1542
+ "loss": 0.4284,
1543
+ "step": 244
1544
+ },
1545
+ {
1546
+ "epoch": 0.34,
1547
+ "learning_rate": 0.0002,
1548
+ "loss": 0.5485,
1549
+ "step": 245
1550
+ },
1551
+ {
1552
+ "epoch": 0.35,
1553
+ "learning_rate": 0.0002,
1554
+ "loss": 0.5654,
1555
+ "step": 246
1556
+ },
1557
+ {
1558
+ "epoch": 0.35,
1559
+ "learning_rate": 0.0002,
1560
+ "loss": 0.5599,
1561
+ "step": 247
1562
+ },
1563
+ {
1564
+ "epoch": 0.35,
1565
+ "learning_rate": 0.0002,
1566
+ "loss": 0.5756,
1567
+ "step": 248
1568
+ },
1569
+ {
1570
+ "epoch": 0.35,
1571
+ "learning_rate": 0.0002,
1572
+ "loss": 0.4219,
1573
+ "step": 249
1574
+ },
1575
+ {
1576
+ "epoch": 0.35,
1577
+ "learning_rate": 0.0002,
1578
+ "loss": 0.7499,
1579
+ "step": 250
1580
+ },
1581
+ {
1582
+ "epoch": 0.35,
1583
+ "learning_rate": 0.0002,
1584
+ "loss": 0.6233,
1585
+ "step": 251
1586
+ },
1587
+ {
1588
+ "epoch": 0.35,
1589
+ "learning_rate": 0.0002,
1590
+ "loss": 0.6872,
1591
+ "step": 252
1592
+ },
1593
+ {
1594
+ "epoch": 0.36,
1595
+ "learning_rate": 0.0002,
1596
+ "loss": 0.8836,
1597
+ "step": 253
1598
+ },
1599
+ {
1600
+ "epoch": 0.36,
1601
+ "learning_rate": 0.0002,
1602
+ "loss": 0.8242,
1603
+ "step": 254
1604
+ },
1605
+ {
1606
+ "epoch": 0.36,
1607
+ "learning_rate": 0.0002,
1608
+ "loss": 0.9424,
1609
+ "step": 255
1610
+ },
1611
+ {
1612
+ "epoch": 0.36,
1613
+ "learning_rate": 0.0002,
1614
+ "loss": 0.6088,
1615
+ "step": 256
1616
+ },
1617
+ {
1618
+ "epoch": 0.36,
1619
+ "learning_rate": 0.0002,
1620
+ "loss": 0.6363,
1621
+ "step": 257
1622
+ },
1623
+ {
1624
+ "epoch": 0.36,
1625
+ "learning_rate": 0.0002,
1626
+ "loss": 0.6582,
1627
+ "step": 258
1628
+ },
1629
+ {
1630
+ "epoch": 0.36,
1631
+ "learning_rate": 0.0002,
1632
+ "loss": 1.0579,
1633
+ "step": 259
1634
+ },
1635
+ {
1636
+ "epoch": 0.37,
1637
+ "learning_rate": 0.0002,
1638
+ "loss": 0.5946,
1639
+ "step": 260
1640
+ },
1641
+ {
1642
+ "epoch": 0.37,
1643
+ "learning_rate": 0.0002,
1644
+ "loss": 0.9046,
1645
+ "step": 261
1646
+ },
1647
+ {
1648
+ "epoch": 0.37,
1649
+ "learning_rate": 0.0002,
1650
+ "loss": 0.6693,
1651
+ "step": 262
1652
+ },
1653
+ {
1654
+ "epoch": 0.37,
1655
+ "learning_rate": 0.0002,
1656
+ "loss": 0.6784,
1657
+ "step": 263
1658
+ },
1659
+ {
1660
+ "epoch": 0.37,
1661
+ "learning_rate": 0.0002,
1662
+ "loss": 0.4804,
1663
+ "step": 264
1664
+ },
1665
+ {
1666
+ "epoch": 0.37,
1667
+ "learning_rate": 0.0002,
1668
+ "loss": 0.5627,
1669
+ "step": 265
1670
+ },
1671
+ {
1672
+ "epoch": 0.37,
1673
+ "learning_rate": 0.0002,
1674
+ "loss": 0.4451,
1675
+ "step": 266
1676
+ },
1677
+ {
1678
+ "epoch": 0.37,
1679
+ "learning_rate": 0.0002,
1680
+ "loss": 0.3804,
1681
+ "step": 267
1682
+ },
1683
+ {
1684
+ "epoch": 0.38,
1685
+ "learning_rate": 0.0002,
1686
+ "loss": 0.4614,
1687
+ "step": 268
1688
+ },
1689
+ {
1690
+ "epoch": 0.38,
1691
+ "learning_rate": 0.0002,
1692
+ "loss": 0.4386,
1693
+ "step": 269
1694
+ },
1695
+ {
1696
+ "epoch": 0.38,
1697
+ "learning_rate": 0.0002,
1698
+ "loss": 0.5054,
1699
+ "step": 270
1700
+ },
1701
+ {
1702
+ "epoch": 0.38,
1703
+ "learning_rate": 0.0002,
1704
+ "loss": 0.3299,
1705
+ "step": 271
1706
+ },
1707
+ {
1708
+ "epoch": 0.38,
1709
+ "learning_rate": 0.0002,
1710
+ "loss": 0.3767,
1711
+ "step": 272
1712
+ },
1713
+ {
1714
+ "epoch": 0.38,
1715
+ "learning_rate": 0.0002,
1716
+ "loss": 0.6474,
1717
+ "step": 273
1718
+ },
1719
+ {
1720
+ "epoch": 0.38,
1721
+ "learning_rate": 0.0002,
1722
+ "loss": 0.5278,
1723
+ "step": 274
1724
+ },
1725
+ {
1726
+ "epoch": 0.39,
1727
+ "learning_rate": 0.0002,
1728
+ "loss": 0.6876,
1729
+ "step": 275
1730
+ },
1731
+ {
1732
+ "epoch": 0.39,
1733
+ "learning_rate": 0.0002,
1734
+ "loss": 0.6743,
1735
+ "step": 276
1736
+ },
1737
+ {
1738
+ "epoch": 0.39,
1739
+ "learning_rate": 0.0002,
1740
+ "loss": 0.6905,
1741
+ "step": 277
1742
+ },
1743
+ {
1744
+ "epoch": 0.39,
1745
+ "learning_rate": 0.0002,
1746
+ "loss": 0.5167,
1747
+ "step": 278
1748
+ },
1749
+ {
1750
+ "epoch": 0.39,
1751
+ "learning_rate": 0.0002,
1752
+ "loss": 0.8943,
1753
+ "step": 279
1754
+ },
1755
+ {
1756
+ "epoch": 0.39,
1757
+ "learning_rate": 0.0002,
1758
+ "loss": 0.5534,
1759
+ "step": 280
1760
+ },
1761
+ {
1762
+ "epoch": 0.39,
1763
+ "learning_rate": 0.0002,
1764
+ "loss": 0.6501,
1765
+ "step": 281
1766
+ },
1767
+ {
1768
+ "epoch": 0.4,
1769
+ "learning_rate": 0.0002,
1770
+ "loss": 0.4966,
1771
+ "step": 282
1772
+ },
1773
+ {
1774
+ "epoch": 0.4,
1775
+ "learning_rate": 0.0002,
1776
+ "loss": 0.5887,
1777
+ "step": 283
1778
+ },
1779
+ {
1780
+ "epoch": 0.4,
1781
+ "learning_rate": 0.0002,
1782
+ "loss": 0.5823,
1783
+ "step": 284
1784
+ },
1785
+ {
1786
+ "epoch": 0.4,
1787
+ "learning_rate": 0.0002,
1788
+ "loss": 0.4327,
1789
+ "step": 285
1790
+ },
1791
+ {
1792
+ "epoch": 0.4,
1793
+ "learning_rate": 0.0002,
1794
+ "loss": 0.6986,
1795
+ "step": 286
1796
+ },
1797
+ {
1798
+ "epoch": 0.4,
1799
+ "learning_rate": 0.0002,
1800
+ "loss": 0.3851,
1801
+ "step": 287
1802
+ },
1803
+ {
1804
+ "epoch": 0.4,
1805
+ "learning_rate": 0.0002,
1806
+ "loss": 0.9356,
1807
+ "step": 288
1808
+ },
1809
+ {
1810
+ "epoch": 0.41,
1811
+ "learning_rate": 0.0002,
1812
+ "loss": 0.4721,
1813
+ "step": 289
1814
+ },
1815
+ {
1816
+ "epoch": 0.41,
1817
+ "learning_rate": 0.0002,
1818
+ "loss": 0.5978,
1819
+ "step": 290
1820
+ },
1821
+ {
1822
+ "epoch": 0.41,
1823
+ "learning_rate": 0.0002,
1824
+ "loss": 0.5783,
1825
+ "step": 291
1826
+ },
1827
+ {
1828
+ "epoch": 0.41,
1829
+ "learning_rate": 0.0002,
1830
+ "loss": 0.5125,
1831
+ "step": 292
1832
+ },
1833
+ {
1834
+ "epoch": 0.41,
1835
+ "learning_rate": 0.0002,
1836
+ "loss": 0.5305,
1837
+ "step": 293
1838
+ },
1839
+ {
1840
+ "epoch": 0.41,
1841
+ "learning_rate": 0.0002,
1842
+ "loss": 0.5606,
1843
+ "step": 294
1844
+ },
1845
+ {
1846
+ "epoch": 0.41,
1847
+ "learning_rate": 0.0002,
1848
+ "loss": 1.4431,
1849
+ "step": 295
1850
+ },
1851
+ {
1852
+ "epoch": 0.42,
1853
+ "learning_rate": 0.0002,
1854
+ "loss": 0.7331,
1855
+ "step": 296
1856
+ },
1857
+ {
1858
+ "epoch": 0.42,
1859
+ "learning_rate": 0.0002,
1860
+ "loss": 0.6763,
1861
+ "step": 297
1862
+ },
1863
+ {
1864
+ "epoch": 0.42,
1865
+ "learning_rate": 0.0002,
1866
+ "loss": 0.4489,
1867
+ "step": 298
1868
+ },
1869
+ {
1870
+ "epoch": 0.42,
1871
+ "learning_rate": 0.0002,
1872
+ "loss": 0.5757,
1873
+ "step": 299
1874
+ },
1875
+ {
1876
+ "epoch": 0.42,
1877
+ "learning_rate": 0.0002,
1878
+ "loss": 0.6899,
1879
+ "step": 300
1880
+ },
1881
+ {
1882
+ "epoch": 0.42,
1883
+ "learning_rate": 0.0002,
1884
+ "loss": 0.5681,
1885
+ "step": 301
1886
+ },
1887
+ {
1888
+ "epoch": 0.42,
1889
+ "learning_rate": 0.0002,
1890
+ "loss": 0.8638,
1891
+ "step": 302
1892
+ },
1893
+ {
1894
+ "epoch": 0.43,
1895
+ "learning_rate": 0.0002,
1896
+ "loss": 0.5007,
1897
+ "step": 303
1898
+ },
1899
+ {
1900
+ "epoch": 0.43,
1901
+ "learning_rate": 0.0002,
1902
+ "loss": 0.4922,
1903
+ "step": 304
1904
+ },
1905
+ {
1906
+ "epoch": 0.43,
1907
+ "learning_rate": 0.0002,
1908
+ "loss": 0.6042,
1909
+ "step": 305
1910
+ },
1911
+ {
1912
+ "epoch": 0.43,
1913
+ "learning_rate": 0.0002,
1914
+ "loss": 0.6398,
1915
+ "step": 306
1916
+ },
1917
+ {
1918
+ "epoch": 0.43,
1919
+ "learning_rate": 0.0002,
1920
+ "loss": 0.6903,
1921
+ "step": 307
1922
+ },
1923
+ {
1924
+ "epoch": 0.43,
1925
+ "learning_rate": 0.0002,
1926
+ "loss": 0.8877,
1927
+ "step": 308
1928
+ },
1929
+ {
1930
+ "epoch": 0.43,
1931
+ "learning_rate": 0.0002,
1932
+ "loss": 0.7088,
1933
+ "step": 309
1934
+ },
1935
+ {
1936
+ "epoch": 0.44,
1937
+ "learning_rate": 0.0002,
1938
+ "loss": 0.6013,
1939
+ "step": 310
1940
+ },
1941
+ {
1942
+ "epoch": 0.44,
1943
+ "learning_rate": 0.0002,
1944
+ "loss": 0.8319,
1945
+ "step": 311
1946
+ },
1947
+ {
1948
+ "epoch": 0.44,
1949
+ "learning_rate": 0.0002,
1950
+ "loss": 0.4064,
1951
+ "step": 312
1952
+ },
1953
+ {
1954
+ "epoch": 0.44,
1955
+ "learning_rate": 0.0002,
1956
+ "loss": 0.6466,
1957
+ "step": 313
1958
+ },
1959
+ {
1960
+ "epoch": 0.44,
1961
+ "learning_rate": 0.0002,
1962
+ "loss": 0.6873,
1963
+ "step": 314
1964
+ },
1965
+ {
1966
+ "epoch": 0.44,
1967
+ "learning_rate": 0.0002,
1968
+ "loss": 0.9556,
1969
+ "step": 315
1970
+ },
1971
+ {
1972
+ "epoch": 0.44,
1973
+ "learning_rate": 0.0002,
1974
+ "loss": 0.4119,
1975
+ "step": 316
1976
+ },
1977
+ {
1978
+ "epoch": 0.45,
1979
+ "learning_rate": 0.0002,
1980
+ "loss": 0.5019,
1981
+ "step": 317
1982
+ },
1983
+ {
1984
+ "epoch": 0.45,
1985
+ "learning_rate": 0.0002,
1986
+ "loss": 0.6108,
1987
+ "step": 318
1988
+ },
1989
+ {
1990
+ "epoch": 0.45,
1991
+ "learning_rate": 0.0002,
1992
+ "loss": 0.5208,
1993
+ "step": 319
1994
+ },
1995
+ {
1996
+ "epoch": 0.45,
1997
+ "learning_rate": 0.0002,
1998
+ "loss": 0.6555,
1999
+ "step": 320
2000
+ },
2001
+ {
2002
+ "epoch": 0.45,
2003
+ "learning_rate": 0.0002,
2004
+ "loss": 0.3786,
2005
+ "step": 321
2006
+ },
2007
+ {
2008
+ "epoch": 0.45,
2009
+ "learning_rate": 0.0002,
2010
+ "loss": 0.7625,
2011
+ "step": 322
2012
+ },
2013
+ {
2014
+ "epoch": 0.45,
2015
+ "learning_rate": 0.0002,
2016
+ "loss": 0.6542,
2017
+ "step": 323
2018
+ },
2019
+ {
2020
+ "epoch": 0.45,
2021
+ "learning_rate": 0.0002,
2022
+ "loss": 0.7737,
2023
+ "step": 324
2024
+ },
2025
+ {
2026
+ "epoch": 0.46,
2027
+ "learning_rate": 0.0002,
2028
+ "loss": 0.6379,
2029
+ "step": 325
2030
+ },
2031
+ {
2032
+ "epoch": 0.46,
2033
+ "learning_rate": 0.0002,
2034
+ "loss": 0.5464,
2035
+ "step": 326
2036
+ },
2037
+ {
2038
+ "epoch": 0.46,
2039
+ "learning_rate": 0.0002,
2040
+ "loss": 0.6817,
2041
+ "step": 327
2042
+ },
2043
+ {
2044
+ "epoch": 0.46,
2045
+ "learning_rate": 0.0002,
2046
+ "loss": 0.4032,
2047
+ "step": 328
2048
+ },
2049
+ {
2050
+ "epoch": 0.46,
2051
+ "learning_rate": 0.0002,
2052
+ "loss": 1.1049,
2053
+ "step": 329
2054
+ },
2055
+ {
2056
+ "epoch": 0.46,
2057
+ "learning_rate": 0.0002,
2058
+ "loss": 0.5744,
2059
+ "step": 330
2060
+ },
2061
+ {
2062
+ "epoch": 0.46,
2063
+ "learning_rate": 0.0002,
2064
+ "loss": 0.3844,
2065
+ "step": 331
2066
+ },
2067
+ {
2068
+ "epoch": 0.47,
2069
+ "learning_rate": 0.0002,
2070
+ "loss": 0.6441,
2071
+ "step": 332
2072
+ },
2073
+ {
2074
+ "epoch": 0.47,
2075
+ "learning_rate": 0.0002,
2076
+ "loss": 0.3441,
2077
+ "step": 333
2078
+ },
2079
+ {
2080
+ "epoch": 0.47,
2081
+ "learning_rate": 0.0002,
2082
+ "loss": 0.6981,
2083
+ "step": 334
2084
+ },
2085
+ {
2086
+ "epoch": 0.47,
2087
+ "learning_rate": 0.0002,
2088
+ "loss": 0.437,
2089
+ "step": 335
2090
+ },
2091
+ {
2092
+ "epoch": 0.47,
2093
+ "learning_rate": 0.0002,
2094
+ "loss": 0.9664,
2095
+ "step": 336
2096
+ },
2097
+ {
2098
+ "epoch": 0.47,
2099
+ "learning_rate": 0.0002,
2100
+ "loss": 0.5144,
2101
+ "step": 337
2102
+ },
2103
+ {
2104
+ "epoch": 0.47,
2105
+ "learning_rate": 0.0002,
2106
+ "loss": 0.6048,
2107
+ "step": 338
2108
+ },
2109
+ {
2110
+ "epoch": 0.48,
2111
+ "learning_rate": 0.0002,
2112
+ "loss": 0.4976,
2113
+ "step": 339
2114
+ },
2115
+ {
2116
+ "epoch": 0.48,
2117
+ "learning_rate": 0.0002,
2118
+ "loss": 0.5221,
2119
+ "step": 340
2120
+ },
2121
+ {
2122
+ "epoch": 0.48,
2123
+ "learning_rate": 0.0002,
2124
+ "loss": 0.4715,
2125
+ "step": 341
2126
+ },
2127
+ {
2128
+ "epoch": 0.48,
2129
+ "learning_rate": 0.0002,
2130
+ "loss": 0.7036,
2131
+ "step": 342
2132
+ },
2133
+ {
2134
+ "epoch": 0.48,
2135
+ "learning_rate": 0.0002,
2136
+ "loss": 0.5815,
2137
+ "step": 343
2138
+ },
2139
+ {
2140
+ "epoch": 0.48,
2141
+ "learning_rate": 0.0002,
2142
+ "loss": 0.4843,
2143
+ "step": 344
2144
+ },
2145
+ {
2146
+ "epoch": 0.48,
2147
+ "learning_rate": 0.0002,
2148
+ "loss": 0.5816,
2149
+ "step": 345
2150
+ },
2151
+ {
2152
+ "epoch": 0.49,
2153
+ "learning_rate": 0.0002,
2154
+ "loss": 0.2778,
2155
+ "step": 346
2156
+ },
2157
+ {
2158
+ "epoch": 0.49,
2159
+ "learning_rate": 0.0002,
2160
+ "loss": 0.5073,
2161
+ "step": 347
2162
+ },
2163
+ {
2164
+ "epoch": 0.49,
2165
+ "learning_rate": 0.0002,
2166
+ "loss": 0.3484,
2167
+ "step": 348
2168
+ },
2169
+ {
2170
+ "epoch": 0.49,
2171
+ "learning_rate": 0.0002,
2172
+ "loss": 0.4671,
2173
+ "step": 349
2174
+ },
2175
+ {
2176
+ "epoch": 0.49,
2177
+ "learning_rate": 0.0002,
2178
+ "loss": 0.5155,
2179
+ "step": 350
2180
+ },
2181
+ {
2182
+ "epoch": 0.49,
2183
+ "learning_rate": 0.0002,
2184
+ "loss": 0.5333,
2185
+ "step": 351
2186
+ },
2187
+ {
2188
+ "epoch": 0.49,
2189
+ "learning_rate": 0.0002,
2190
+ "loss": 0.7035,
2191
+ "step": 352
2192
+ },
2193
+ {
2194
+ "epoch": 0.5,
2195
+ "learning_rate": 0.0002,
2196
+ "loss": 0.6029,
2197
+ "step": 353
2198
+ },
2199
+ {
2200
+ "epoch": 0.5,
2201
+ "learning_rate": 0.0002,
2202
+ "loss": 0.9084,
2203
+ "step": 354
2204
+ },
2205
+ {
2206
+ "epoch": 0.5,
2207
+ "learning_rate": 0.0002,
2208
+ "loss": 0.6098,
2209
+ "step": 355
2210
+ },
2211
+ {
2212
+ "epoch": 0.5,
2213
+ "learning_rate": 0.0002,
2214
+ "loss": 0.3604,
2215
+ "step": 356
2216
+ },
2217
+ {
2218
+ "epoch": 0.5,
2219
+ "learning_rate": 0.0002,
2220
+ "loss": 0.5221,
2221
+ "step": 357
2222
+ },
2223
+ {
2224
+ "epoch": 0.5,
2225
+ "learning_rate": 0.0002,
2226
+ "loss": 0.4751,
2227
+ "step": 358
2228
+ },
2229
+ {
2230
+ "epoch": 0.5,
2231
+ "learning_rate": 0.0002,
2232
+ "loss": 0.9367,
2233
+ "step": 359
2234
+ },
2235
+ {
2236
+ "epoch": 0.51,
2237
+ "learning_rate": 0.0002,
2238
+ "loss": 0.418,
2239
+ "step": 360
2240
+ },
2241
+ {
2242
+ "epoch": 0.51,
2243
+ "learning_rate": 0.0002,
2244
+ "loss": 0.3642,
2245
+ "step": 361
2246
+ },
2247
+ {
2248
+ "epoch": 0.51,
2249
+ "learning_rate": 0.0002,
2250
+ "loss": 0.5618,
2251
+ "step": 362
2252
+ },
2253
+ {
2254
+ "epoch": 0.51,
2255
+ "learning_rate": 0.0002,
2256
+ "loss": 0.5638,
2257
+ "step": 363
2258
+ },
2259
+ {
2260
+ "epoch": 0.51,
2261
+ "learning_rate": 0.0002,
2262
+ "loss": 0.9389,
2263
+ "step": 364
2264
+ },
2265
+ {
2266
+ "epoch": 0.51,
2267
+ "learning_rate": 0.0002,
2268
+ "loss": 0.8087,
2269
+ "step": 365
2270
+ },
2271
+ {
2272
+ "epoch": 0.51,
2273
+ "learning_rate": 0.0002,
2274
+ "loss": 0.5299,
2275
+ "step": 366
2276
+ },
2277
+ {
2278
+ "epoch": 0.52,
2279
+ "learning_rate": 0.0002,
2280
+ "loss": 0.8063,
2281
+ "step": 367
2282
+ },
2283
+ {
2284
+ "epoch": 0.52,
2285
+ "learning_rate": 0.0002,
2286
+ "loss": 0.4529,
2287
+ "step": 368
2288
+ },
2289
+ {
2290
+ "epoch": 0.52,
2291
+ "learning_rate": 0.0002,
2292
+ "loss": 0.6113,
2293
+ "step": 369
2294
+ },
2295
+ {
2296
+ "epoch": 0.52,
2297
+ "learning_rate": 0.0002,
2298
+ "loss": 0.813,
2299
+ "step": 370
2300
+ },
2301
+ {
2302
+ "epoch": 0.52,
2303
+ "learning_rate": 0.0002,
2304
+ "loss": 0.7885,
2305
+ "step": 371
2306
+ },
2307
+ {
2308
+ "epoch": 0.52,
2309
+ "learning_rate": 0.0002,
2310
+ "loss": 0.4745,
2311
+ "step": 372
2312
+ },
2313
+ {
2314
+ "epoch": 0.52,
2315
+ "learning_rate": 0.0002,
2316
+ "loss": 0.7722,
2317
+ "step": 373
2318
+ },
2319
+ {
2320
+ "epoch": 0.53,
2321
+ "learning_rate": 0.0002,
2322
+ "loss": 0.7242,
2323
+ "step": 374
2324
+ },
2325
+ {
2326
+ "epoch": 0.53,
2327
+ "learning_rate": 0.0002,
2328
+ "loss": 0.4531,
2329
+ "step": 375
2330
+ },
2331
+ {
2332
+ "epoch": 0.53,
2333
+ "learning_rate": 0.0002,
2334
+ "loss": 0.9054,
2335
+ "step": 376
2336
+ },
2337
+ {
2338
+ "epoch": 0.53,
2339
+ "learning_rate": 0.0002,
2340
+ "loss": 0.9152,
2341
+ "step": 377
2342
+ },
2343
+ {
2344
+ "epoch": 0.53,
2345
+ "learning_rate": 0.0002,
2346
+ "loss": 0.4414,
2347
+ "step": 378
2348
+ },
2349
+ {
2350
+ "epoch": 0.53,
2351
+ "learning_rate": 0.0002,
2352
+ "loss": 0.5136,
2353
+ "step": 379
2354
+ },
2355
+ {
2356
+ "epoch": 0.53,
2357
+ "learning_rate": 0.0002,
2358
+ "loss": 0.3315,
2359
+ "step": 380
2360
+ },
2361
+ {
2362
+ "epoch": 0.53,
2363
+ "learning_rate": 0.0002,
2364
+ "loss": 0.5251,
2365
+ "step": 381
2366
+ },
2367
+ {
2368
+ "epoch": 0.54,
2369
+ "learning_rate": 0.0002,
2370
+ "loss": 0.7332,
2371
+ "step": 382
2372
+ },
2373
+ {
2374
+ "epoch": 0.54,
2375
+ "learning_rate": 0.0002,
2376
+ "loss": 0.5032,
2377
+ "step": 383
2378
+ },
2379
+ {
2380
+ "epoch": 0.54,
2381
+ "learning_rate": 0.0002,
2382
+ "loss": 0.5535,
2383
+ "step": 384
2384
+ },
2385
+ {
2386
+ "epoch": 0.54,
2387
+ "learning_rate": 0.0002,
2388
+ "loss": 0.5916,
2389
+ "step": 385
2390
+ },
2391
+ {
2392
+ "epoch": 0.54,
2393
+ "learning_rate": 0.0002,
2394
+ "loss": 0.6259,
2395
+ "step": 386
2396
+ },
2397
+ {
2398
+ "epoch": 0.54,
2399
+ "learning_rate": 0.0002,
2400
+ "loss": 0.6234,
2401
+ "step": 387
2402
+ },
2403
+ {
2404
+ "epoch": 0.54,
2405
+ "learning_rate": 0.0002,
2406
+ "loss": 0.5661,
2407
+ "step": 388
2408
+ },
2409
+ {
2410
+ "epoch": 0.55,
2411
+ "learning_rate": 0.0002,
2412
+ "loss": 0.3717,
2413
+ "step": 389
2414
+ },
2415
+ {
2416
+ "epoch": 0.55,
2417
+ "learning_rate": 0.0002,
2418
+ "loss": 0.4362,
2419
+ "step": 390
2420
+ },
2421
+ {
2422
+ "epoch": 0.55,
2423
+ "learning_rate": 0.0002,
2424
+ "loss": 0.6232,
2425
+ "step": 391
2426
+ },
2427
+ {
2428
+ "epoch": 0.55,
2429
+ "learning_rate": 0.0002,
2430
+ "loss": 0.6514,
2431
+ "step": 392
2432
+ },
2433
+ {
2434
+ "epoch": 0.55,
2435
+ "learning_rate": 0.0002,
2436
+ "loss": 1.2381,
2437
+ "step": 393
2438
+ },
2439
+ {
2440
+ "epoch": 0.55,
2441
+ "learning_rate": 0.0002,
2442
+ "loss": 0.599,
2443
+ "step": 394
2444
+ },
2445
+ {
2446
+ "epoch": 0.55,
2447
+ "learning_rate": 0.0002,
2448
+ "loss": 0.5788,
2449
+ "step": 395
2450
+ },
2451
+ {
2452
+ "epoch": 0.56,
2453
+ "learning_rate": 0.0002,
2454
+ "loss": 0.4485,
2455
+ "step": 396
2456
+ },
2457
+ {
2458
+ "epoch": 0.56,
2459
+ "learning_rate": 0.0002,
2460
+ "loss": 0.5755,
2461
+ "step": 397
2462
+ },
2463
+ {
2464
+ "epoch": 0.56,
2465
+ "learning_rate": 0.0002,
2466
+ "loss": 0.7999,
2467
+ "step": 398
2468
+ },
2469
+ {
2470
+ "epoch": 0.56,
2471
+ "learning_rate": 0.0002,
2472
+ "loss": 0.4548,
2473
+ "step": 399
2474
+ },
2475
+ {
2476
+ "epoch": 0.56,
2477
+ "learning_rate": 0.0002,
2478
+ "loss": 0.5872,
2479
+ "step": 400
2480
+ },
2481
+ {
2482
+ "epoch": 0.56,
2483
+ "eval_loss": 0.5880547761917114,
2484
+ "eval_runtime": 396.9467,
2485
+ "eval_samples_per_second": 2.519,
2486
+ "eval_steps_per_second": 1.26,
2487
+ "step": 400
2488
+ },
2489
+ {
2490
+ "epoch": 0.56,
2491
+ "mmlu_eval_accuracy": 0.4846205369111515,
2492
+ "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
2493
+ "mmlu_eval_accuracy_anatomy": 0.35714285714285715,
2494
+ "mmlu_eval_accuracy_astronomy": 0.4375,
2495
+ "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
2496
+ "mmlu_eval_accuracy_clinical_knowledge": 0.41379310344827586,
2497
+ "mmlu_eval_accuracy_college_biology": 0.4375,
2498
+ "mmlu_eval_accuracy_college_chemistry": 0.125,
2499
+ "mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
2500
+ "mmlu_eval_accuracy_college_mathematics": 0.36363636363636365,
2501
+ "mmlu_eval_accuracy_college_medicine": 0.36363636363636365,
2502
+ "mmlu_eval_accuracy_college_physics": 0.2727272727272727,
2503
+ "mmlu_eval_accuracy_computer_security": 0.45454545454545453,
2504
+ "mmlu_eval_accuracy_conceptual_physics": 0.38461538461538464,
2505
+ "mmlu_eval_accuracy_econometrics": 0.3333333333333333,
2506
+ "mmlu_eval_accuracy_electrical_engineering": 0.5625,
2507
+ "mmlu_eval_accuracy_elementary_mathematics": 0.3170731707317073,
2508
+ "mmlu_eval_accuracy_formal_logic": 0.35714285714285715,
2509
+ "mmlu_eval_accuracy_global_facts": 0.2,
2510
+ "mmlu_eval_accuracy_high_school_biology": 0.28125,
2511
+ "mmlu_eval_accuracy_high_school_chemistry": 0.2727272727272727,
2512
+ "mmlu_eval_accuracy_high_school_computer_science": 0.7777777777777778,
2513
+ "mmlu_eval_accuracy_high_school_european_history": 0.6111111111111112,
2514
+ "mmlu_eval_accuracy_high_school_geography": 0.7727272727272727,
2515
+ "mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
2516
+ "mmlu_eval_accuracy_high_school_macroeconomics": 0.32558139534883723,
2517
+ "mmlu_eval_accuracy_high_school_mathematics": 0.4482758620689655,
2518
+ "mmlu_eval_accuracy_high_school_microeconomics": 0.34615384615384615,
2519
+ "mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
2520
+ "mmlu_eval_accuracy_high_school_psychology": 0.75,
2521
+ "mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
2522
+ "mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
2523
+ "mmlu_eval_accuracy_high_school_world_history": 0.5769230769230769,
2524
+ "mmlu_eval_accuracy_human_aging": 0.6956521739130435,
2525
+ "mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
2526
+ "mmlu_eval_accuracy_international_law": 0.6923076923076923,
2527
+ "mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
2528
+ "mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
2529
+ "mmlu_eval_accuracy_machine_learning": 0.45454545454545453,
2530
+ "mmlu_eval_accuracy_management": 0.7272727272727273,
2531
+ "mmlu_eval_accuracy_marketing": 0.76,
2532
+ "mmlu_eval_accuracy_medical_genetics": 0.8181818181818182,
2533
+ "mmlu_eval_accuracy_miscellaneous": 0.6395348837209303,
2534
+ "mmlu_eval_accuracy_moral_disputes": 0.39473684210526316,
2535
+ "mmlu_eval_accuracy_moral_scenarios": 0.25,
2536
+ "mmlu_eval_accuracy_nutrition": 0.5454545454545454,
2537
+ "mmlu_eval_accuracy_philosophy": 0.6176470588235294,
2538
+ "mmlu_eval_accuracy_prehistory": 0.45714285714285713,
2539
+ "mmlu_eval_accuracy_professional_accounting": 0.41935483870967744,
2540
+ "mmlu_eval_accuracy_professional_law": 0.25882352941176473,
2541
+ "mmlu_eval_accuracy_professional_medicine": 0.45161290322580644,
2542
+ "mmlu_eval_accuracy_professional_psychology": 0.43478260869565216,
2543
+ "mmlu_eval_accuracy_public_relations": 0.5,
2544
+ "mmlu_eval_accuracy_security_studies": 0.48148148148148145,
2545
+ "mmlu_eval_accuracy_sociology": 0.6818181818181818,
2546
+ "mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
2547
+ "mmlu_eval_accuracy_virology": 0.5,
2548
+ "mmlu_eval_accuracy_world_religions": 0.6842105263157895,
2549
+ "mmlu_loss": 0.9576842238246305,
2550
+ "step": 400
2551
+ },
2552
+ {
2553
+ "epoch": 0.56,
2554
+ "learning_rate": 0.0002,
2555
+ "loss": 0.3718,
2556
+ "step": 401
2557
+ },
2558
+ {
2559
+ "epoch": 0.56,
2560
+ "learning_rate": 0.0002,
2561
+ "loss": 0.5159,
2562
+ "step": 402
2563
+ },
2564
+ {
2565
+ "epoch": 0.57,
2566
+ "learning_rate": 0.0002,
2567
+ "loss": 0.7165,
2568
+ "step": 403
2569
+ },
2570
+ {
2571
+ "epoch": 0.57,
2572
+ "learning_rate": 0.0002,
2573
+ "loss": 0.6522,
2574
+ "step": 404
2575
+ },
2576
+ {
2577
+ "epoch": 0.57,
2578
+ "learning_rate": 0.0002,
2579
+ "loss": 0.4325,
2580
+ "step": 405
2581
+ },
2582
+ {
2583
+ "epoch": 0.57,
2584
+ "learning_rate": 0.0002,
2585
+ "loss": 0.5079,
2586
+ "step": 406
2587
+ },
2588
+ {
2589
+ "epoch": 0.57,
2590
+ "learning_rate": 0.0002,
2591
+ "loss": 0.6084,
2592
+ "step": 407
2593
+ },
2594
+ {
2595
+ "epoch": 0.57,
2596
+ "learning_rate": 0.0002,
2597
+ "loss": 0.8387,
2598
+ "step": 408
2599
+ },
2600
+ {
2601
+ "epoch": 0.57,
2602
+ "learning_rate": 0.0002,
2603
+ "loss": 0.3619,
2604
+ "step": 409
2605
+ },
2606
+ {
2607
+ "epoch": 0.58,
2608
+ "learning_rate": 0.0002,
2609
+ "loss": 0.4273,
2610
+ "step": 410
2611
+ },
2612
+ {
2613
+ "epoch": 0.58,
2614
+ "learning_rate": 0.0002,
2615
+ "loss": 0.4658,
2616
+ "step": 411
2617
+ },
2618
+ {
2619
+ "epoch": 0.58,
2620
+ "learning_rate": 0.0002,
2621
+ "loss": 0.4996,
2622
+ "step": 412
2623
+ },
2624
+ {
2625
+ "epoch": 0.58,
2626
+ "learning_rate": 0.0002,
2627
+ "loss": 0.6387,
2628
+ "step": 413
2629
+ },
2630
+ {
2631
+ "epoch": 0.58,
2632
+ "learning_rate": 0.0002,
2633
+ "loss": 0.7449,
2634
+ "step": 414
2635
+ },
2636
+ {
2637
+ "epoch": 0.58,
2638
+ "learning_rate": 0.0002,
2639
+ "loss": 0.5566,
2640
+ "step": 415
2641
+ },
2642
+ {
2643
+ "epoch": 0.58,
2644
+ "learning_rate": 0.0002,
2645
+ "loss": 0.6642,
2646
+ "step": 416
2647
+ },
2648
+ {
2649
+ "epoch": 0.59,
2650
+ "learning_rate": 0.0002,
2651
+ "loss": 0.8422,
2652
+ "step": 417
2653
+ },
2654
+ {
2655
+ "epoch": 0.59,
2656
+ "learning_rate": 0.0002,
2657
+ "loss": 0.9808,
2658
+ "step": 418
2659
+ },
2660
+ {
2661
+ "epoch": 0.59,
2662
+ "learning_rate": 0.0002,
2663
+ "loss": 1.0236,
2664
+ "step": 419
2665
+ },
2666
+ {
2667
+ "epoch": 0.59,
2668
+ "learning_rate": 0.0002,
2669
+ "loss": 0.8144,
2670
+ "step": 420
2671
+ },
2672
+ {
2673
+ "epoch": 0.59,
2674
+ "learning_rate": 0.0002,
2675
+ "loss": 0.5159,
2676
+ "step": 421
2677
+ },
2678
+ {
2679
+ "epoch": 0.59,
2680
+ "learning_rate": 0.0002,
2681
+ "loss": 0.5916,
2682
+ "step": 422
2683
+ },
2684
+ {
2685
+ "epoch": 0.59,
2686
+ "learning_rate": 0.0002,
2687
+ "loss": 0.5595,
2688
+ "step": 423
2689
+ },
2690
+ {
2691
+ "epoch": 0.6,
2692
+ "learning_rate": 0.0002,
2693
+ "loss": 0.5962,
2694
+ "step": 424
2695
+ },
2696
+ {
2697
+ "epoch": 0.6,
2698
+ "learning_rate": 0.0002,
2699
+ "loss": 0.7472,
2700
+ "step": 425
2701
+ },
2702
+ {
2703
+ "epoch": 0.6,
2704
+ "learning_rate": 0.0002,
2705
+ "loss": 0.3524,
2706
+ "step": 426
2707
+ },
2708
+ {
2709
+ "epoch": 0.6,
2710
+ "learning_rate": 0.0002,
2711
+ "loss": 0.4166,
2712
+ "step": 427
2713
+ },
2714
+ {
2715
+ "epoch": 0.6,
2716
+ "learning_rate": 0.0002,
2717
+ "loss": 0.7061,
2718
+ "step": 428
2719
+ },
2720
+ {
2721
+ "epoch": 0.6,
2722
+ "learning_rate": 0.0002,
2723
+ "loss": 0.5893,
2724
+ "step": 429
2725
+ },
2726
+ {
2727
+ "epoch": 0.6,
2728
+ "learning_rate": 0.0002,
2729
+ "loss": 0.5269,
2730
+ "step": 430
2731
+ },
2732
+ {
2733
+ "epoch": 0.61,
2734
+ "learning_rate": 0.0002,
2735
+ "loss": 0.43,
2736
+ "step": 431
2737
+ },
2738
+ {
2739
+ "epoch": 0.61,
2740
+ "learning_rate": 0.0002,
2741
+ "loss": 0.6153,
2742
+ "step": 432
2743
+ },
2744
+ {
2745
+ "epoch": 0.61,
2746
+ "learning_rate": 0.0002,
2747
+ "loss": 0.6293,
2748
+ "step": 433
2749
+ },
2750
+ {
2751
+ "epoch": 0.61,
2752
+ "learning_rate": 0.0002,
2753
+ "loss": 0.4561,
2754
+ "step": 434
2755
+ },
2756
+ {
2757
+ "epoch": 0.61,
2758
+ "learning_rate": 0.0002,
2759
+ "loss": 0.9173,
2760
+ "step": 435
2761
+ },
2762
+ {
2763
+ "epoch": 0.61,
2764
+ "learning_rate": 0.0002,
2765
+ "loss": 0.2933,
2766
+ "step": 436
2767
+ },
2768
+ {
2769
+ "epoch": 0.61,
2770
+ "learning_rate": 0.0002,
2771
+ "loss": 0.7743,
2772
+ "step": 437
2773
+ },
2774
+ {
2775
+ "epoch": 0.61,
2776
+ "learning_rate": 0.0002,
2777
+ "loss": 0.8235,
2778
+ "step": 438
2779
+ },
2780
+ {
2781
+ "epoch": 0.62,
2782
+ "learning_rate": 0.0002,
2783
+ "loss": 0.5366,
2784
+ "step": 439
2785
+ },
2786
+ {
2787
+ "epoch": 0.62,
2788
+ "learning_rate": 0.0002,
2789
+ "loss": 0.4424,
2790
+ "step": 440
2791
+ },
2792
+ {
2793
+ "epoch": 0.62,
2794
+ "learning_rate": 0.0002,
2795
+ "loss": 0.7654,
2796
+ "step": 441
2797
+ },
2798
+ {
2799
+ "epoch": 0.62,
2800
+ "learning_rate": 0.0002,
2801
+ "loss": 0.6101,
2802
+ "step": 442
2803
+ },
2804
+ {
2805
+ "epoch": 0.62,
2806
+ "learning_rate": 0.0002,
2807
+ "loss": 0.4104,
2808
+ "step": 443
2809
+ },
2810
+ {
2811
+ "epoch": 0.62,
2812
+ "learning_rate": 0.0002,
2813
+ "loss": 0.7698,
2814
+ "step": 444
2815
+ },
2816
+ {
2817
+ "epoch": 0.62,
2818
+ "learning_rate": 0.0002,
2819
+ "loss": 0.521,
2820
+ "step": 445
2821
+ },
2822
+ {
2823
+ "epoch": 0.63,
2824
+ "learning_rate": 0.0002,
2825
+ "loss": 0.6492,
2826
+ "step": 446
2827
+ },
2828
+ {
2829
+ "epoch": 0.63,
2830
+ "learning_rate": 0.0002,
2831
+ "loss": 0.6252,
2832
+ "step": 447
2833
+ },
2834
+ {
2835
+ "epoch": 0.63,
2836
+ "learning_rate": 0.0002,
2837
+ "loss": 0.6444,
2838
+ "step": 448
2839
+ },
2840
+ {
2841
+ "epoch": 0.63,
2842
+ "learning_rate": 0.0002,
2843
+ "loss": 0.9372,
2844
+ "step": 449
2845
+ },
2846
+ {
2847
+ "epoch": 0.63,
2848
+ "learning_rate": 0.0002,
2849
+ "loss": 0.5931,
2850
+ "step": 450
2851
+ },
2852
+ {
2853
+ "epoch": 0.63,
2854
+ "learning_rate": 0.0002,
2855
+ "loss": 0.581,
2856
+ "step": 451
2857
+ },
2858
+ {
2859
+ "epoch": 0.63,
2860
+ "learning_rate": 0.0002,
2861
+ "loss": 0.7696,
2862
+ "step": 452
2863
+ },
2864
+ {
2865
+ "epoch": 0.64,
2866
+ "learning_rate": 0.0002,
2867
+ "loss": 0.6884,
2868
+ "step": 453
2869
+ },
2870
+ {
2871
+ "epoch": 0.64,
2872
+ "learning_rate": 0.0002,
2873
+ "loss": 0.4991,
2874
+ "step": 454
2875
+ },
2876
+ {
2877
+ "epoch": 0.64,
2878
+ "learning_rate": 0.0002,
2879
+ "loss": 0.3674,
2880
+ "step": 455
2881
+ },
2882
+ {
2883
+ "epoch": 0.64,
2884
+ "learning_rate": 0.0002,
2885
+ "loss": 0.4275,
2886
+ "step": 456
2887
+ },
2888
+ {
2889
+ "epoch": 0.64,
2890
+ "learning_rate": 0.0002,
2891
+ "loss": 0.6982,
2892
+ "step": 457
2893
+ },
2894
+ {
2895
+ "epoch": 0.64,
2896
+ "learning_rate": 0.0002,
2897
+ "loss": 0.6223,
2898
+ "step": 458
2899
+ },
2900
+ {
2901
+ "epoch": 0.64,
2902
+ "learning_rate": 0.0002,
2903
+ "loss": 0.5234,
2904
+ "step": 459
2905
+ },
2906
+ {
2907
+ "epoch": 0.65,
2908
+ "learning_rate": 0.0002,
2909
+ "loss": 0.8479,
2910
+ "step": 460
2911
+ },
2912
+ {
2913
+ "epoch": 0.65,
2914
+ "learning_rate": 0.0002,
2915
+ "loss": 0.5331,
2916
+ "step": 461
2917
+ },
2918
+ {
2919
+ "epoch": 0.65,
2920
+ "learning_rate": 0.0002,
2921
+ "loss": 0.5743,
2922
+ "step": 462
2923
+ },
2924
+ {
2925
+ "epoch": 0.65,
2926
+ "learning_rate": 0.0002,
2927
+ "loss": 0.6911,
2928
+ "step": 463
2929
+ },
2930
+ {
2931
+ "epoch": 0.65,
2932
+ "learning_rate": 0.0002,
2933
+ "loss": 0.4317,
2934
+ "step": 464
2935
+ },
2936
+ {
2937
+ "epoch": 0.65,
2938
+ "learning_rate": 0.0002,
2939
+ "loss": 0.4714,
2940
+ "step": 465
2941
+ },
2942
+ {
2943
+ "epoch": 0.65,
2944
+ "learning_rate": 0.0002,
2945
+ "loss": 0.7314,
2946
+ "step": 466
2947
+ },
2948
+ {
2949
+ "epoch": 0.66,
2950
+ "learning_rate": 0.0002,
2951
+ "loss": 0.6088,
2952
+ "step": 467
2953
+ },
2954
+ {
2955
+ "epoch": 0.66,
2956
+ "learning_rate": 0.0002,
2957
+ "loss": 0.8494,
2958
+ "step": 468
2959
+ },
2960
+ {
2961
+ "epoch": 0.66,
2962
+ "learning_rate": 0.0002,
2963
+ "loss": 0.7006,
2964
+ "step": 469
2965
+ },
2966
+ {
2967
+ "epoch": 0.66,
2968
+ "learning_rate": 0.0002,
2969
+ "loss": 0.4734,
2970
+ "step": 470
2971
+ },
2972
+ {
2973
+ "epoch": 0.66,
2974
+ "learning_rate": 0.0002,
2975
+ "loss": 0.6942,
2976
+ "step": 471
2977
+ },
2978
+ {
2979
+ "epoch": 0.66,
2980
+ "learning_rate": 0.0002,
2981
+ "loss": 0.6707,
2982
+ "step": 472
2983
+ },
2984
+ {
2985
+ "epoch": 0.66,
2986
+ "learning_rate": 0.0002,
2987
+ "loss": 0.4737,
2988
+ "step": 473
2989
+ },
2990
+ {
2991
+ "epoch": 0.67,
2992
+ "learning_rate": 0.0002,
2993
+ "loss": 0.4141,
2994
+ "step": 474
2995
+ },
2996
+ {
2997
+ "epoch": 0.67,
2998
+ "learning_rate": 0.0002,
2999
+ "loss": 0.4816,
3000
+ "step": 475
3001
+ },
3002
+ {
3003
+ "epoch": 0.67,
3004
+ "learning_rate": 0.0002,
3005
+ "loss": 0.6768,
3006
+ "step": 476
3007
+ },
3008
+ {
3009
+ "epoch": 0.67,
3010
+ "learning_rate": 0.0002,
3011
+ "loss": 0.4494,
3012
+ "step": 477
3013
+ },
3014
+ {
3015
+ "epoch": 0.67,
3016
+ "learning_rate": 0.0002,
3017
+ "loss": 0.3933,
3018
+ "step": 478
3019
+ },
3020
+ {
3021
+ "epoch": 0.67,
3022
+ "learning_rate": 0.0002,
3023
+ "loss": 0.3092,
3024
+ "step": 479
3025
+ },
3026
+ {
3027
+ "epoch": 0.67,
3028
+ "learning_rate": 0.0002,
3029
+ "loss": 0.5429,
3030
+ "step": 480
3031
+ },
3032
+ {
3033
+ "epoch": 0.68,
3034
+ "learning_rate": 0.0002,
3035
+ "loss": 0.5985,
3036
+ "step": 481
3037
+ },
3038
+ {
3039
+ "epoch": 0.68,
3040
+ "learning_rate": 0.0002,
3041
+ "loss": 0.658,
3042
+ "step": 482
3043
+ },
3044
+ {
3045
+ "epoch": 0.68,
3046
+ "learning_rate": 0.0002,
3047
+ "loss": 0.5346,
3048
+ "step": 483
3049
+ },
3050
+ {
3051
+ "epoch": 0.68,
3052
+ "learning_rate": 0.0002,
3053
+ "loss": 0.2877,
3054
+ "step": 484
3055
+ },
3056
+ {
3057
+ "epoch": 0.68,
3058
+ "learning_rate": 0.0002,
3059
+ "loss": 0.9522,
3060
+ "step": 485
3061
+ },
3062
+ {
3063
+ "epoch": 0.68,
3064
+ "learning_rate": 0.0002,
3065
+ "loss": 0.2777,
3066
+ "step": 486
3067
+ },
3068
+ {
3069
+ "epoch": 0.68,
3070
+ "learning_rate": 0.0002,
3071
+ "loss": 0.5888,
3072
+ "step": 487
3073
+ },
3074
+ {
3075
+ "epoch": 0.69,
3076
+ "learning_rate": 0.0002,
3077
+ "loss": 0.5125,
3078
+ "step": 488
3079
+ },
3080
+ {
3081
+ "epoch": 0.69,
3082
+ "learning_rate": 0.0002,
3083
+ "loss": 0.597,
3084
+ "step": 489
3085
+ },
3086
+ {
3087
+ "epoch": 0.69,
3088
+ "learning_rate": 0.0002,
3089
+ "loss": 0.6547,
3090
+ "step": 490
3091
+ },
3092
+ {
3093
+ "epoch": 0.69,
3094
+ "learning_rate": 0.0002,
3095
+ "loss": 0.4975,
3096
+ "step": 491
3097
+ },
3098
+ {
3099
+ "epoch": 0.69,
3100
+ "learning_rate": 0.0002,
3101
+ "loss": 0.556,
3102
+ "step": 492
3103
+ },
3104
+ {
3105
+ "epoch": 0.69,
3106
+ "learning_rate": 0.0002,
3107
+ "loss": 0.6408,
3108
+ "step": 493
3109
+ },
3110
+ {
3111
+ "epoch": 0.69,
3112
+ "learning_rate": 0.0002,
3113
+ "loss": 0.8527,
3114
+ "step": 494
3115
+ },
3116
+ {
3117
+ "epoch": 0.69,
3118
+ "learning_rate": 0.0002,
3119
+ "loss": 0.5432,
3120
+ "step": 495
3121
+ },
3122
+ {
3123
+ "epoch": 0.7,
3124
+ "learning_rate": 0.0002,
3125
+ "loss": 0.6178,
3126
+ "step": 496
3127
+ },
3128
+ {
3129
+ "epoch": 0.7,
3130
+ "learning_rate": 0.0002,
3131
+ "loss": 0.8967,
3132
+ "step": 497
3133
+ },
3134
+ {
3135
+ "epoch": 0.7,
3136
+ "learning_rate": 0.0002,
3137
+ "loss": 0.5481,
3138
+ "step": 498
3139
+ },
3140
+ {
3141
+ "epoch": 0.7,
3142
+ "learning_rate": 0.0002,
3143
+ "loss": 0.4747,
3144
+ "step": 499
3145
+ },
3146
+ {
3147
+ "epoch": 0.7,
3148
+ "learning_rate": 0.0002,
3149
+ "loss": 0.522,
3150
+ "step": 500
3151
+ },
3152
+ {
3153
+ "epoch": 0.7,
3154
+ "learning_rate": 0.0002,
3155
+ "loss": 0.2957,
3156
+ "step": 501
3157
+ },
3158
+ {
3159
+ "epoch": 0.7,
3160
+ "learning_rate": 0.0002,
3161
+ "loss": 0.5323,
3162
+ "step": 502
3163
+ },
3164
+ {
3165
+ "epoch": 0.71,
3166
+ "learning_rate": 0.0002,
3167
+ "loss": 0.3375,
3168
+ "step": 503
3169
+ },
3170
+ {
3171
+ "epoch": 0.71,
3172
+ "learning_rate": 0.0002,
3173
+ "loss": 0.4292,
3174
+ "step": 504
3175
+ },
3176
+ {
3177
+ "epoch": 0.71,
3178
+ "learning_rate": 0.0002,
3179
+ "loss": 0.375,
3180
+ "step": 505
3181
+ },
3182
+ {
3183
+ "epoch": 0.71,
3184
+ "learning_rate": 0.0002,
3185
+ "loss": 0.4781,
3186
+ "step": 506
3187
+ },
3188
+ {
3189
+ "epoch": 0.71,
3190
+ "learning_rate": 0.0002,
3191
+ "loss": 0.3998,
3192
+ "step": 507
3193
+ },
3194
+ {
3195
+ "epoch": 0.71,
3196
+ "learning_rate": 0.0002,
3197
+ "loss": 0.4019,
3198
+ "step": 508
3199
+ },
3200
+ {
3201
+ "epoch": 0.71,
3202
+ "learning_rate": 0.0002,
3203
+ "loss": 0.5622,
3204
+ "step": 509
3205
+ },
3206
+ {
3207
+ "epoch": 0.72,
3208
+ "learning_rate": 0.0002,
3209
+ "loss": 0.5633,
3210
+ "step": 510
3211
+ },
3212
+ {
3213
+ "epoch": 0.72,
3214
+ "learning_rate": 0.0002,
3215
+ "loss": 0.6693,
3216
+ "step": 511
3217
+ },
3218
+ {
3219
+ "epoch": 0.72,
3220
+ "learning_rate": 0.0002,
3221
+ "loss": 0.4402,
3222
+ "step": 512
3223
+ },
3224
+ {
3225
+ "epoch": 0.72,
3226
+ "learning_rate": 0.0002,
3227
+ "loss": 1.0258,
3228
+ "step": 513
3229
+ },
3230
+ {
3231
+ "epoch": 0.72,
3232
+ "learning_rate": 0.0002,
3233
+ "loss": 0.5866,
3234
+ "step": 514
3235
+ },
3236
+ {
3237
+ "epoch": 0.72,
3238
+ "learning_rate": 0.0002,
3239
+ "loss": 0.738,
3240
+ "step": 515
3241
+ },
3242
+ {
3243
+ "epoch": 0.72,
3244
+ "learning_rate": 0.0002,
3245
+ "loss": 0.4765,
3246
+ "step": 516
3247
+ },
3248
+ {
3249
+ "epoch": 0.73,
3250
+ "learning_rate": 0.0002,
3251
+ "loss": 0.748,
3252
+ "step": 517
3253
+ },
3254
+ {
3255
+ "epoch": 0.73,
3256
+ "learning_rate": 0.0002,
3257
+ "loss": 0.6294,
3258
+ "step": 518
3259
+ },
3260
+ {
3261
+ "epoch": 0.73,
3262
+ "learning_rate": 0.0002,
3263
+ "loss": 0.7203,
3264
+ "step": 519
3265
+ },
3266
+ {
3267
+ "epoch": 0.73,
3268
+ "learning_rate": 0.0002,
3269
+ "loss": 0.5033,
3270
+ "step": 520
3271
+ },
3272
+ {
3273
+ "epoch": 0.73,
3274
+ "learning_rate": 0.0002,
3275
+ "loss": 0.7007,
3276
+ "step": 521
3277
+ },
3278
+ {
3279
+ "epoch": 0.73,
3280
+ "learning_rate": 0.0002,
3281
+ "loss": 0.5197,
3282
+ "step": 522
3283
+ },
3284
+ {
3285
+ "epoch": 0.73,
3286
+ "learning_rate": 0.0002,
3287
+ "loss": 0.8251,
3288
+ "step": 523
3289
+ },
3290
+ {
3291
+ "epoch": 0.74,
3292
+ "learning_rate": 0.0002,
3293
+ "loss": 0.6264,
3294
+ "step": 524
3295
+ },
3296
+ {
3297
+ "epoch": 0.74,
3298
+ "learning_rate": 0.0002,
3299
+ "loss": 0.422,
3300
+ "step": 525
3301
+ },
3302
+ {
3303
+ "epoch": 0.74,
3304
+ "learning_rate": 0.0002,
3305
+ "loss": 0.3288,
3306
+ "step": 526
3307
+ },
3308
+ {
3309
+ "epoch": 0.74,
3310
+ "learning_rate": 0.0002,
3311
+ "loss": 0.7641,
3312
+ "step": 527
3313
+ },
3314
+ {
3315
+ "epoch": 0.74,
3316
+ "learning_rate": 0.0002,
3317
+ "loss": 0.8521,
3318
+ "step": 528
3319
+ },
3320
+ {
3321
+ "epoch": 0.74,
3322
+ "learning_rate": 0.0002,
3323
+ "loss": 0.3265,
3324
+ "step": 529
3325
+ },
3326
+ {
3327
+ "epoch": 0.74,
3328
+ "learning_rate": 0.0002,
3329
+ "loss": 0.6372,
3330
+ "step": 530
3331
+ },
3332
+ {
3333
+ "epoch": 0.75,
3334
+ "learning_rate": 0.0002,
3335
+ "loss": 0.5141,
3336
+ "step": 531
3337
+ },
3338
+ {
3339
+ "epoch": 0.75,
3340
+ "learning_rate": 0.0002,
3341
+ "loss": 0.5788,
3342
+ "step": 532
3343
+ },
3344
+ {
3345
+ "epoch": 0.75,
3346
+ "learning_rate": 0.0002,
3347
+ "loss": 0.6268,
3348
+ "step": 533
3349
+ },
3350
+ {
3351
+ "epoch": 0.75,
3352
+ "learning_rate": 0.0002,
3353
+ "loss": 0.3942,
3354
+ "step": 534
3355
+ },
3356
+ {
3357
+ "epoch": 0.75,
3358
+ "learning_rate": 0.0002,
3359
+ "loss": 0.5028,
3360
+ "step": 535
3361
+ },
3362
+ {
3363
+ "epoch": 0.75,
3364
+ "learning_rate": 0.0002,
3365
+ "loss": 0.6894,
3366
+ "step": 536
3367
+ },
3368
+ {
3369
+ "epoch": 0.75,
3370
+ "learning_rate": 0.0002,
3371
+ "loss": 0.5062,
3372
+ "step": 537
3373
+ },
3374
+ {
3375
+ "epoch": 0.76,
3376
+ "learning_rate": 0.0002,
3377
+ "loss": 0.3966,
3378
+ "step": 538
3379
+ },
3380
+ {
3381
+ "epoch": 0.76,
3382
+ "learning_rate": 0.0002,
3383
+ "loss": 0.5734,
3384
+ "step": 539
3385
+ },
3386
+ {
3387
+ "epoch": 0.76,
3388
+ "learning_rate": 0.0002,
3389
+ "loss": 0.67,
3390
+ "step": 540
3391
+ },
3392
+ {
3393
+ "epoch": 0.76,
3394
+ "learning_rate": 0.0002,
3395
+ "loss": 0.6711,
3396
+ "step": 541
3397
+ },
3398
+ {
3399
+ "epoch": 0.76,
3400
+ "learning_rate": 0.0002,
3401
+ "loss": 0.8185,
3402
+ "step": 542
3403
+ },
3404
+ {
3405
+ "epoch": 0.76,
3406
+ "learning_rate": 0.0002,
3407
+ "loss": 0.6515,
3408
+ "step": 543
3409
+ },
3410
+ {
3411
+ "epoch": 0.76,
3412
+ "learning_rate": 0.0002,
3413
+ "loss": 0.4948,
3414
+ "step": 544
3415
+ },
3416
+ {
3417
+ "epoch": 0.77,
3418
+ "learning_rate": 0.0002,
3419
+ "loss": 0.5231,
3420
+ "step": 545
3421
+ },
3422
+ {
3423
+ "epoch": 0.77,
3424
+ "learning_rate": 0.0002,
3425
+ "loss": 0.5238,
3426
+ "step": 546
3427
+ },
3428
+ {
3429
+ "epoch": 0.77,
3430
+ "learning_rate": 0.0002,
3431
+ "loss": 0.4768,
3432
+ "step": 547
3433
+ },
3434
+ {
3435
+ "epoch": 0.77,
3436
+ "learning_rate": 0.0002,
3437
+ "loss": 0.5402,
3438
+ "step": 548
3439
+ },
3440
+ {
3441
+ "epoch": 0.77,
3442
+ "learning_rate": 0.0002,
3443
+ "loss": 0.4775,
3444
+ "step": 549
3445
+ },
3446
+ {
3447
+ "epoch": 0.77,
3448
+ "learning_rate": 0.0002,
3449
+ "loss": 0.7488,
3450
+ "step": 550
3451
+ },
3452
+ {
3453
+ "epoch": 0.77,
3454
+ "learning_rate": 0.0002,
3455
+ "loss": 0.4362,
3456
+ "step": 551
3457
+ },
3458
+ {
3459
+ "epoch": 0.78,
3460
+ "learning_rate": 0.0002,
3461
+ "loss": 0.4829,
3462
+ "step": 552
3463
+ },
3464
+ {
3465
+ "epoch": 0.78,
3466
+ "learning_rate": 0.0002,
3467
+ "loss": 0.5964,
3468
+ "step": 553
3469
+ },
3470
+ {
3471
+ "epoch": 0.78,
3472
+ "learning_rate": 0.0002,
3473
+ "loss": 0.6663,
3474
+ "step": 554
3475
+ },
3476
+ {
3477
+ "epoch": 0.78,
3478
+ "learning_rate": 0.0002,
3479
+ "loss": 0.7621,
3480
+ "step": 555
3481
+ },
3482
+ {
3483
+ "epoch": 0.78,
3484
+ "learning_rate": 0.0002,
3485
+ "loss": 0.4115,
3486
+ "step": 556
3487
+ },
3488
+ {
3489
+ "epoch": 0.78,
3490
+ "learning_rate": 0.0002,
3491
+ "loss": 0.6349,
3492
+ "step": 557
3493
+ },
3494
+ {
3495
+ "epoch": 0.78,
3496
+ "learning_rate": 0.0002,
3497
+ "loss": 0.6089,
3498
+ "step": 558
3499
+ },
3500
+ {
3501
+ "epoch": 0.78,
3502
+ "learning_rate": 0.0002,
3503
+ "loss": 0.4772,
3504
+ "step": 559
3505
+ },
3506
+ {
3507
+ "epoch": 0.79,
3508
+ "learning_rate": 0.0002,
3509
+ "loss": 0.5788,
3510
+ "step": 560
3511
+ },
3512
+ {
3513
+ "epoch": 0.79,
3514
+ "learning_rate": 0.0002,
3515
+ "loss": 0.3905,
3516
+ "step": 561
3517
+ },
3518
+ {
3519
+ "epoch": 0.79,
3520
+ "learning_rate": 0.0002,
3521
+ "loss": 0.509,
3522
+ "step": 562
3523
+ },
3524
+ {
3525
+ "epoch": 0.79,
3526
+ "learning_rate": 0.0002,
3527
+ "loss": 0.355,
3528
+ "step": 563
3529
+ },
3530
+ {
3531
+ "epoch": 0.79,
3532
+ "learning_rate": 0.0002,
3533
+ "loss": 0.7796,
3534
+ "step": 564
3535
+ },
3536
+ {
3537
+ "epoch": 0.79,
3538
+ "learning_rate": 0.0002,
3539
+ "loss": 0.6693,
3540
+ "step": 565
3541
+ },
3542
+ {
3543
+ "epoch": 0.79,
3544
+ "learning_rate": 0.0002,
3545
+ "loss": 0.6663,
3546
+ "step": 566
3547
+ },
3548
+ {
3549
+ "epoch": 0.8,
3550
+ "learning_rate": 0.0002,
3551
+ "loss": 0.5275,
3552
+ "step": 567
3553
+ },
3554
+ {
3555
+ "epoch": 0.8,
3556
+ "learning_rate": 0.0002,
3557
+ "loss": 0.8837,
3558
+ "step": 568
3559
+ },
3560
+ {
3561
+ "epoch": 0.8,
3562
+ "learning_rate": 0.0002,
3563
+ "loss": 0.4286,
3564
+ "step": 569
3565
+ },
3566
+ {
3567
+ "epoch": 0.8,
3568
+ "learning_rate": 0.0002,
3569
+ "loss": 1.0252,
3570
+ "step": 570
3571
+ },
3572
+ {
3573
+ "epoch": 0.8,
3574
+ "learning_rate": 0.0002,
3575
+ "loss": 0.5022,
3576
+ "step": 571
3577
+ },
3578
+ {
3579
+ "epoch": 0.8,
3580
+ "learning_rate": 0.0002,
3581
+ "loss": 0.6754,
3582
+ "step": 572
3583
+ },
3584
+ {
3585
+ "epoch": 0.8,
3586
+ "learning_rate": 0.0002,
3587
+ "loss": 0.4,
3588
+ "step": 573
3589
+ },
3590
+ {
3591
+ "epoch": 0.81,
3592
+ "learning_rate": 0.0002,
3593
+ "loss": 0.7761,
3594
+ "step": 574
3595
+ },
3596
+ {
3597
+ "epoch": 0.81,
3598
+ "learning_rate": 0.0002,
3599
+ "loss": 0.6813,
3600
+ "step": 575
3601
+ },
3602
+ {
3603
+ "epoch": 0.81,
3604
+ "learning_rate": 0.0002,
3605
+ "loss": 0.5806,
3606
+ "step": 576
3607
+ },
3608
+ {
3609
+ "epoch": 0.81,
3610
+ "learning_rate": 0.0002,
3611
+ "loss": 0.5755,
3612
+ "step": 577
3613
+ },
3614
+ {
3615
+ "epoch": 0.81,
3616
+ "learning_rate": 0.0002,
3617
+ "loss": 0.7661,
3618
+ "step": 578
3619
+ },
3620
+ {
3621
+ "epoch": 0.81,
3622
+ "learning_rate": 0.0002,
3623
+ "loss": 0.8814,
3624
+ "step": 579
3625
+ },
3626
+ {
3627
+ "epoch": 0.81,
3628
+ "learning_rate": 0.0002,
3629
+ "loss": 0.5416,
3630
+ "step": 580
3631
+ },
3632
+ {
3633
+ "epoch": 0.82,
3634
+ "learning_rate": 0.0002,
3635
+ "loss": 0.4392,
3636
+ "step": 581
3637
+ },
3638
+ {
3639
+ "epoch": 0.82,
3640
+ "learning_rate": 0.0002,
3641
+ "loss": 0.5973,
3642
+ "step": 582
3643
+ },
3644
+ {
3645
+ "epoch": 0.82,
3646
+ "learning_rate": 0.0002,
3647
+ "loss": 0.4373,
3648
+ "step": 583
3649
+ },
3650
+ {
3651
+ "epoch": 0.82,
3652
+ "learning_rate": 0.0002,
3653
+ "loss": 0.5291,
3654
+ "step": 584
3655
+ },
3656
+ {
3657
+ "epoch": 0.82,
3658
+ "learning_rate": 0.0002,
3659
+ "loss": 0.485,
3660
+ "step": 585
3661
+ },
3662
+ {
3663
+ "epoch": 0.82,
3664
+ "learning_rate": 0.0002,
3665
+ "loss": 0.353,
3666
+ "step": 586
3667
+ },
3668
+ {
3669
+ "epoch": 0.82,
3670
+ "learning_rate": 0.0002,
3671
+ "loss": 0.9446,
3672
+ "step": 587
3673
+ },
3674
+ {
3675
+ "epoch": 0.83,
3676
+ "learning_rate": 0.0002,
3677
+ "loss": 0.4602,
3678
+ "step": 588
3679
+ },
3680
+ {
3681
+ "epoch": 0.83,
3682
+ "learning_rate": 0.0002,
3683
+ "loss": 0.7173,
3684
+ "step": 589
3685
+ },
3686
+ {
3687
+ "epoch": 0.83,
3688
+ "learning_rate": 0.0002,
3689
+ "loss": 0.4454,
3690
+ "step": 590
3691
+ },
3692
+ {
3693
+ "epoch": 0.83,
3694
+ "learning_rate": 0.0002,
3695
+ "loss": 0.4104,
3696
+ "step": 591
3697
+ },
3698
+ {
3699
+ "epoch": 0.83,
3700
+ "learning_rate": 0.0002,
3701
+ "loss": 0.3425,
3702
+ "step": 592
3703
+ },
3704
+ {
3705
+ "epoch": 0.83,
3706
+ "learning_rate": 0.0002,
3707
+ "loss": 0.4651,
3708
+ "step": 593
3709
+ },
3710
+ {
3711
+ "epoch": 0.83,
3712
+ "learning_rate": 0.0002,
3713
+ "loss": 0.6563,
3714
+ "step": 594
3715
+ },
3716
+ {
3717
+ "epoch": 0.84,
3718
+ "learning_rate": 0.0002,
3719
+ "loss": 0.5574,
3720
+ "step": 595
3721
+ },
3722
+ {
3723
+ "epoch": 0.84,
3724
+ "learning_rate": 0.0002,
3725
+ "loss": 0.4641,
3726
+ "step": 596
3727
+ },
3728
+ {
3729
+ "epoch": 0.84,
3730
+ "learning_rate": 0.0002,
3731
+ "loss": 0.576,
3732
+ "step": 597
3733
+ },
3734
+ {
3735
+ "epoch": 0.84,
3736
+ "learning_rate": 0.0002,
3737
+ "loss": 0.4755,
3738
+ "step": 598
3739
+ },
3740
+ {
3741
+ "epoch": 0.84,
3742
+ "learning_rate": 0.0002,
3743
+ "loss": 0.7452,
3744
+ "step": 599
3745
+ },
3746
+ {
3747
+ "epoch": 0.84,
3748
+ "learning_rate": 0.0002,
3749
+ "loss": 0.6964,
3750
+ "step": 600
3751
+ },
3752
+ {
3753
+ "epoch": 0.84,
3754
+ "eval_loss": 0.5618544220924377,
3755
+ "eval_runtime": 397.2406,
3756
+ "eval_samples_per_second": 2.517,
3757
+ "eval_steps_per_second": 1.259,
3758
+ "step": 600
3759
+ },
3760
+ {
3761
+ "epoch": 0.84,
3762
+ "mmlu_eval_accuracy": 0.5192617436513831,
3763
+ "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
3764
+ "mmlu_eval_accuracy_anatomy": 0.42857142857142855,
3765
+ "mmlu_eval_accuracy_astronomy": 0.5,
3766
+ "mmlu_eval_accuracy_business_ethics": 0.45454545454545453,
3767
+ "mmlu_eval_accuracy_clinical_knowledge": 0.41379310344827586,
3768
+ "mmlu_eval_accuracy_college_biology": 0.5,
3769
+ "mmlu_eval_accuracy_college_chemistry": 0.125,
3770
+ "mmlu_eval_accuracy_college_computer_science": 0.7272727272727273,
3771
+ "mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
3772
+ "mmlu_eval_accuracy_college_medicine": 0.5,
3773
+ "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
3774
+ "mmlu_eval_accuracy_computer_security": 0.5454545454545454,
3775
+ "mmlu_eval_accuracy_conceptual_physics": 0.34615384615384615,
3776
+ "mmlu_eval_accuracy_econometrics": 0.25,
3777
+ "mmlu_eval_accuracy_electrical_engineering": 0.4375,
3778
+ "mmlu_eval_accuracy_elementary_mathematics": 0.4146341463414634,
3779
+ "mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
3780
+ "mmlu_eval_accuracy_global_facts": 0.2,
3781
+ "mmlu_eval_accuracy_high_school_biology": 0.375,
3782
+ "mmlu_eval_accuracy_high_school_chemistry": 0.3181818181818182,
3783
+ "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
3784
+ "mmlu_eval_accuracy_high_school_european_history": 0.6666666666666666,
3785
+ "mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
3786
+ "mmlu_eval_accuracy_high_school_government_and_politics": 0.8095238095238095,
3787
+ "mmlu_eval_accuracy_high_school_macroeconomics": 0.4186046511627907,
3788
+ "mmlu_eval_accuracy_high_school_mathematics": 0.41379310344827586,
3789
+ "mmlu_eval_accuracy_high_school_microeconomics": 0.5,
3790
+ "mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
3791
+ "mmlu_eval_accuracy_high_school_psychology": 0.7666666666666667,
3792
+ "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
3793
+ "mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
3794
+ "mmlu_eval_accuracy_high_school_world_history": 0.6153846153846154,
3795
+ "mmlu_eval_accuracy_human_aging": 0.6086956521739131,
3796
+ "mmlu_eval_accuracy_human_sexuality": 0.5,
3797
+ "mmlu_eval_accuracy_international_law": 0.6923076923076923,
3798
+ "mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
3799
+ "mmlu_eval_accuracy_logical_fallacies": 0.7777777777777778,
3800
+ "mmlu_eval_accuracy_machine_learning": 0.6363636363636364,
3801
+ "mmlu_eval_accuracy_management": 0.9090909090909091,
3802
+ "mmlu_eval_accuracy_marketing": 0.72,
3803
+ "mmlu_eval_accuracy_medical_genetics": 0.8181818181818182,
3804
+ "mmlu_eval_accuracy_miscellaneous": 0.6511627906976745,
3805
+ "mmlu_eval_accuracy_moral_disputes": 0.5,
3806
+ "mmlu_eval_accuracy_moral_scenarios": 0.28,
3807
+ "mmlu_eval_accuracy_nutrition": 0.696969696969697,
3808
+ "mmlu_eval_accuracy_philosophy": 0.5294117647058824,
3809
+ "mmlu_eval_accuracy_prehistory": 0.4857142857142857,
3810
+ "mmlu_eval_accuracy_professional_accounting": 0.41935483870967744,
3811
+ "mmlu_eval_accuracy_professional_law": 0.4235294117647059,
3812
+ "mmlu_eval_accuracy_professional_medicine": 0.41935483870967744,
3813
+ "mmlu_eval_accuracy_professional_psychology": 0.5072463768115942,
3814
+ "mmlu_eval_accuracy_public_relations": 0.5,
3815
+ "mmlu_eval_accuracy_security_studies": 0.48148148148148145,
3816
+ "mmlu_eval_accuracy_sociology": 0.7727272727272727,
3817
+ "mmlu_eval_accuracy_us_foreign_policy": 0.6363636363636364,
3818
+ "mmlu_eval_accuracy_virology": 0.5555555555555556,
3819
+ "mmlu_eval_accuracy_world_religions": 0.7368421052631579,
3820
+ "mmlu_loss": 0.7100075650522509,
3821
+ "step": 600
3822
+ }
3823
+ ],
3824
+ "max_steps": 5000,
3825
+ "num_train_epochs": 8,
3826
+ "total_flos": 3.241392402011259e+17,
3827
+ "trial_name": null,
3828
+ "trial_params": null
3829
+ }
checkpoint-600/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85d09a9f54aa026a0295bc7d334655520e9bf512c90be29e98b4ed206613fba6
3
+ size 5947