Hugging Face's logo

Siddeshwar1625
/
osint-checkpoints

Safetensors
Model card Files
xet
 Community
Siddeshwar1625 commited on
Commit
0f12618
·
verified ·
1 Parent(s): 9e67e28

Intermediate checkpoint upload step=60 (generator_train)

Browse files
Files changed (12) hide show
  1. .gitattributes +1 -0
  2. self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/chat_template.jinja +54 -0
  3. self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/config.json +57 -0
  4. self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/generation_config.json +13 -0
  5. self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/model.safetensors +3 -0
  6. self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/optimizer.pt +3 -0
  7. self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/rng_state.pth +3 -0
  8. self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/scheduler.pt +3 -0
  9. self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/tokenizer.json +3 -0
  10. self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/tokenizer_config.json +32 -0
  11. self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/trainer_state.json +1129 -0
  12. self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/training_args.bin +3 -0
.gitattributes CHANGED
@@ -73,3 +73,4 @@ self_play_hf_l40s_full/round_003/answerer_train/checkpoint-90/tokenizer.json fil
73
  self_play_hf_l40s_full/round_003/answerer_train/checkpoint-120/tokenizer.json filter=lfs diff=lfs merge=lfs -text
74
  self_play_hf_l40s_full/round_003/answerer_train/final_model/tokenizer.json filter=lfs diff=lfs merge=lfs -text
75
  self_play_hf_l40s_full/round_004/generator_train/checkpoint-30/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
73
  self_play_hf_l40s_full/round_003/answerer_train/checkpoint-120/tokenizer.json filter=lfs diff=lfs merge=lfs -text
74
  self_play_hf_l40s_full/round_003/answerer_train/final_model/tokenizer.json filter=lfs diff=lfs merge=lfs -text
75
  self_play_hf_l40s_full/round_004/generator_train/checkpoint-30/tokenizer.json filter=lfs diff=lfs merge=lfs -text
76
+ self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/tokenizer.json filter=lfs diff=lfs merge=lfs -text
self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen2ForCausalLM"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "bos_token_id": null,
7
+ "dtype": "float32",
8
+ "eos_token_id": 151645,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 896,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 4864,
13
+ "layer_types": [
14
+ "full_attention",
15
+ "full_attention",
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention"
38
+ ],
39
+ "max_position_embeddings": 32768,
40
+ "max_window_layers": 21,
41
+ "model_type": "qwen2",
42
+ "num_attention_heads": 14,
43
+ "num_hidden_layers": 24,
44
+ "num_key_value_heads": 2,
45
+ "pad_token_id": 151643,
46
+ "rms_norm_eps": 1e-06,
47
+ "rope_parameters": {
48
+ "rope_theta": 1000000.0,
49
+ "rope_type": "default"
50
+ },
51
+ "sliding_window": null,
52
+ "tie_word_embeddings": true,
53
+ "transformers_version": "5.6.2",
54
+ "use_cache": false,
55
+ "use_sliding_window": false,
56
+ "vocab_size": 151936
57
+ }
self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_sample": true,
3
+ "eos_token_id": [
4
+ 151645,
5
+ 151643
6
+ ],
7
+ "pad_token_id": 151643,
8
+ "repetition_penalty": 1.1,
9
+ "temperature": 0.7,
10
+ "top_k": 20,
11
+ "top_p": 0.8,
12
+ "transformers_version": "5.6.2"
13
+ }
self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0bd4d8b0f12cca5935397910860ea9289bda7d47c7cb6a1dd9de303cda0638ac
3
+ size 1976163472
self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4897da33f4371838492266c818960ca6c9100505d2739940b5c375686f1d0517
3
+ size 3952509771
self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72d73b1faaa85ee872c45feb745fc0ce8498b5608889164df261d7d0fd5b01a4
3
+ size 14645
self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ef717a3e0187059e90f9fce4130d81b5008cd97668718926e21d89d9ccfe2ac
3
+ size 1465
self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8
3
+ size 11421892
self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/tokenizer_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": null,
5
+ "clean_up_tokenization_spaces": false,
6
+ "eos_token": "<|im_end|>",
7
+ "errors": "replace",
8
+ "extra_special_tokens": [
9
+ "<|im_start|>",
10
+ "<|im_end|>",
11
+ "<|object_ref_start|>",
12
+ "<|object_ref_end|>",
13
+ "<|box_start|>",
14
+ "<|box_end|>",
15
+ "<|quad_start|>",
16
+ "<|quad_end|>",
17
+ "<|vision_start|>",
18
+ "<|vision_end|>",
19
+ "<|vision_pad|>",
20
+ "<|image_pad|>",
21
+ "<|video_pad|>"
22
+ ],
23
+ "is_local": true,
24
+ "local_files_only": false,
25
+ "model_max_length": 131072,
26
+ "pad_token": "<|endoftext|>",
27
+ "padding_side": "left",
28
+ "split_special_tokens": false,
29
+ "tokenizer_class": "Qwen2Tokenizer",
30
+ "truncation_side": "left",
31
+ "unk_token": null
32
+ }
self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/trainer_state.json ADDED
@@ -0,0 +1,1129 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 2.5,
6
+ "eval_steps": 500,
7
+ "global_step": 60,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "clip_ratio/high_max": 0.0,
14
+ "clip_ratio/high_mean": 0.0,
15
+ "clip_ratio/low_mean": 0.0,
16
+ "clip_ratio/low_min": 0.0,
17
+ "clip_ratio/region_mean": 0.0,
18
+ "completions/clipped_ratio": 1.0,
19
+ "completions/max_length": 384.0,
20
+ "completions/max_terminated_length": 0.0,
21
+ "completions/mean_length": 384.0,
22
+ "completions/mean_terminated_length": 0.0,
23
+ "completions/min_length": 384.0,
24
+ "completions/min_terminated_length": 0.0,
25
+ "entropy": 3.7540836334228516,
26
+ "epoch": 0.041666666666666664,
27
+ "frac_reward_zero_std": 0.5,
28
+ "grad_norm": 2.3061139583587646,
29
+ "kl": 0.0004934151656925678,
30
+ "learning_rate": 5e-06,
31
+ "loss": 0.062387701123952866,
32
+ "num_tokens": 25136.0,
33
+ "reward": -0.054375000298023224,
34
+ "reward_std": 0.01249999925494194,
35
+ "rewards/GeneratorRewardFunction/mean": -0.054375000298023224,
36
+ "rewards/GeneratorRewardFunction/std": 0.012500000186264515,
37
+ "step": 1,
38
+ "step_time": 12.28940931599982
39
+ },
40
+ {
41
+ "clip_ratio/high_max": 0.03515625,
42
+ "clip_ratio/high_mean": 0.03515625,
43
+ "clip_ratio/low_mean": 0.0,
44
+ "clip_ratio/low_min": 0.0,
45
+ "clip_ratio/region_mean": 0.03515625,
46
+ "entropy": 3.8097915649414062,
47
+ "epoch": 0.08333333333333333,
48
+ "grad_norm": 1.1211391687393188,
49
+ "kl": 0.017926404252648354,
50
+ "learning_rate": 4.958333333333334e-06,
51
+ "loss": -0.05972687527537346,
52
+ "step": 2,
53
+ "step_time": 0.22016475100008392
54
+ },
55
+ {
56
+ "clip_ratio/high_max": 0.0,
57
+ "clip_ratio/high_mean": 0.0,
58
+ "clip_ratio/low_mean": 0.0559895820915699,
59
+ "clip_ratio/low_min": 0.0559895820915699,
60
+ "clip_ratio/region_mean": 0.0559895820915699,
61
+ "entropy": 3.656827688217163,
62
+ "epoch": 0.125,
63
+ "grad_norm": 2.1082754135131836,
64
+ "kl": 0.025636015459895134,
65
+ "learning_rate": 4.9166666666666665e-06,
66
+ "loss": 0.09194447845220566,
67
+ "step": 3,
68
+ "step_time": 0.21996182600014436
69
+ },
70
+ {
71
+ "clip_ratio/high_max": 0.0963541641831398,
72
+ "clip_ratio/high_mean": 0.0963541641831398,
73
+ "clip_ratio/low_mean": 0.0,
74
+ "clip_ratio/low_min": 0.0,
75
+ "clip_ratio/region_mean": 0.0963541641831398,
76
+ "entropy": 3.8841819763183594,
77
+ "epoch": 0.16666666666666666,
78
+ "grad_norm": 1.4164149761199951,
79
+ "kl": 0.030918851494789124,
80
+ "learning_rate": 4.875e-06,
81
+ "loss": -0.08998001366853714,
82
+ "step": 4,
83
+ "step_time": 0.22261608200005867
84
+ },
85
+ {
86
+ "clip_ratio/high_max": 0.0,
87
+ "clip_ratio/high_mean": 0.0,
88
+ "clip_ratio/low_mean": 0.0,
89
+ "clip_ratio/low_min": 0.0,
90
+ "clip_ratio/region_mean": 0.0,
91
+ "completions/clipped_ratio": 1.0,
92
+ "completions/max_length": 384.0,
93
+ "completions/max_terminated_length": 0.0,
94
+ "completions/mean_length": 384.0,
95
+ "completions/mean_terminated_length": 0.0,
96
+ "completions/min_length": 384.0,
97
+ "completions/min_terminated_length": 0.0,
98
+ "entropy": 3.838679075241089,
99
+ "epoch": 0.20833333333333334,
100
+ "frac_reward_zero_std": 1.0,
101
+ "grad_norm": 0.019757986068725586,
102
+ "kl": 0.038326989859342575,
103
+ "learning_rate": 4.833333333333333e-06,
104
+ "loss": 9.581747144693509e-05,
105
+ "num_tokens": 50436.0,
106
+ "reward": -0.05000000074505806,
107
+ "reward_std": 0.0,
108
+ "rewards/GeneratorRewardFunction/mean": -0.05000000074505806,
109
+ "rewards/GeneratorRewardFunction/std": 0.0,
110
+ "step": 5,
111
+ "step_time": 12.358392488999925
112
+ },
113
+ {
114
+ "clip_ratio/high_max": 0.0,
115
+ "clip_ratio/high_mean": 0.0,
116
+ "clip_ratio/low_mean": 0.0,
117
+ "clip_ratio/low_min": 0.0,
118
+ "clip_ratio/region_mean": 0.0,
119
+ "entropy": 3.878758192062378,
120
+ "epoch": 0.25,
121
+ "grad_norm": 0.02407441847026348,
122
+ "kl": 0.047756802290678024,
123
+ "learning_rate": 4.791666666666668e-06,
124
+ "loss": 0.00011939200339838862,
125
+ "step": 6,
126
+ "step_time": 0.22772415899999032
127
+ },
128
+ {
129
+ "clip_ratio/high_max": 0.0,
130
+ "clip_ratio/high_mean": 0.0,
131
+ "clip_ratio/low_mean": 0.0,
132
+ "clip_ratio/low_min": 0.0,
133
+ "clip_ratio/region_mean": 0.0,
134
+ "entropy": 4.379356861114502,
135
+ "epoch": 0.2916666666666667,
136
+ "grad_norm": 0.021465642377734184,
137
+ "kl": 0.04809901490807533,
138
+ "learning_rate": 4.75e-06,
139
+ "loss": 0.00012024753959849477,
140
+ "step": 7,
141
+ "step_time": 0.22733985799959555
142
+ },
143
+ {
144
+ "clip_ratio/high_max": 0.0,
145
+ "clip_ratio/high_mean": 0.0,
146
+ "clip_ratio/low_mean": 0.0,
147
+ "clip_ratio/low_min": 0.0,
148
+ "clip_ratio/region_mean": 0.0,
149
+ "entropy": 4.279653072357178,
150
+ "epoch": 0.3333333333333333,
151
+ "grad_norm": 0.0189784225076437,
152
+ "kl": 0.04601867124438286,
153
+ "learning_rate": 4.708333333333334e-06,
154
+ "loss": 0.00011504666326800361,
155
+ "step": 8,
156
+ "step_time": 0.22832407399982912
157
+ },
158
+ {
159
+ "clip_ratio/high_max": 0.0,
160
+ "clip_ratio/high_mean": 0.0,
161
+ "clip_ratio/low_mean": 0.0,
162
+ "clip_ratio/low_min": 0.0,
163
+ "clip_ratio/region_mean": 0.0,
164
+ "completions/clipped_ratio": 1.0,
165
+ "completions/max_length": 384.0,
166
+ "completions/max_terminated_length": 0.0,
167
+ "completions/mean_length": 384.0,
168
+ "completions/mean_terminated_length": 0.0,
169
+ "completions/min_length": 384.0,
170
+ "completions/min_terminated_length": 0.0,
171
+ "entropy": 3.9055023193359375,
172
+ "epoch": 0.375,
173
+ "frac_reward_zero_std": 0.5,
174
+ "grad_norm": 2.340376377105713,
175
+ "kl": 0.06852344423532486,
176
+ "learning_rate": 4.666666666666667e-06,
177
+ "loss": 0.09388399124145508,
178
+ "num_tokens": 75808.0,
179
+ "reward": -0.11156250536441803,
180
+ "reward_std": 0.16717973351478577,
181
+ "rewards/GeneratorRewardFunction/mean": -0.11156250536441803,
182
+ "rewards/GeneratorRewardFunction/std": 0.16717973351478577,
183
+ "step": 9,
184
+ "step_time": 12.377079298000353
185
+ },
186
+ {
187
+ "clip_ratio/high_max": 0.0013020833721384406,
188
+ "clip_ratio/high_mean": 0.0013020833721384406,
189
+ "clip_ratio/low_mean": 0.001953125,
190
+ "clip_ratio/low_min": 0.001953125,
191
+ "clip_ratio/region_mean": 0.0032552082557231188,
192
+ "entropy": 4.169831275939941,
193
+ "epoch": 0.4166666666666667,
194
+ "grad_norm": 2.4143083095550537,
195
+ "kl": 0.08672171831130981,
196
+ "learning_rate": 4.625000000000001e-06,
197
+ "loss": 0.06073353812098503,
198
+ "step": 10,
199
+ "step_time": 0.2275676320000457
200
+ },
201
+ {
202
+ "clip_ratio/high_max": 0.017578125,
203
+ "clip_ratio/high_mean": 0.017578125,
204
+ "clip_ratio/low_mean": 0.0,
205
+ "clip_ratio/low_min": 0.0,
206
+ "clip_ratio/region_mean": 0.017578125,
207
+ "entropy": 4.554050445556641,
208
+ "epoch": 0.4583333333333333,
209
+ "grad_norm": 1.0698223114013672,
210
+ "kl": 0.07464811205863953,
211
+ "learning_rate": 4.583333333333333e-06,
212
+ "loss": -0.06039423868060112,
213
+ "step": 11,
214
+ "step_time": 0.22664018099976602
215
+ },
216
+ {
217
+ "clip_ratio/high_max": 0.060546875,
218
+ "clip_ratio/high_mean": 0.060546875,
219
+ "clip_ratio/low_mean": 0.0,
220
+ "clip_ratio/low_min": 0.0,
221
+ "clip_ratio/region_mean": 0.060546875,
222
+ "entropy": 4.123767375946045,
223
+ "epoch": 0.5,
224
+ "grad_norm": 1.3299524784088135,
225
+ "kl": 0.11542084068059921,
226
+ "learning_rate": 4.541666666666667e-06,
227
+ "loss": -0.09344702959060669,
228
+ "step": 12,
229
+ "step_time": 0.22850085099980788
230
+ },
231
+ {
232
+ "clip_ratio/high_max": 0.0,
233
+ "clip_ratio/high_mean": 0.0,
234
+ "clip_ratio/low_mean": 0.0,
235
+ "clip_ratio/low_min": 0.0,
236
+ "clip_ratio/region_mean": 0.0,
237
+ "completions/clipped_ratio": 1.0,
238
+ "completions/max_length": 384.0,
239
+ "completions/max_terminated_length": 0.0,
240
+ "completions/mean_length": 384.0,
241
+ "completions/mean_terminated_length": 0.0,
242
+ "completions/min_length": 384.0,
243
+ "completions/min_terminated_length": 0.0,
244
+ "entropy": 4.361082553863525,
245
+ "epoch": 0.5416666666666666,
246
+ "frac_reward_zero_std": 0.5,
247
+ "grad_norm": 2.413485288619995,
248
+ "kl": 0.09952286630868912,
249
+ "learning_rate": 4.5e-06,
250
+ "loss": 0.062085848301649094,
251
+ "num_tokens": 100880.0,
252
+ "reward": -0.08906249701976776,
253
+ "reward_std": 0.14971326291561127,
254
+ "rewards/GeneratorRewardFunction/mean": -0.08906249701976776,
255
+ "rewards/GeneratorRewardFunction/std": 0.14971326291561127,
256
+ "step": 13,
257
+ "step_time": 12.362482152999746
258
+ },
259
+ {
260
+ "clip_ratio/high_max": 0.0045572915114462376,
261
+ "clip_ratio/high_mean": 0.0045572915114462376,
262
+ "clip_ratio/low_mean": 0.0013020833721384406,
263
+ "clip_ratio/low_min": 0.0013020833721384406,
264
+ "clip_ratio/region_mean": 0.005859375,
265
+ "entropy": 4.488595485687256,
266
+ "epoch": 0.5833333333333334,
267
+ "grad_norm": 2.6962649822235107,
268
+ "kl": 0.10717153549194336,
269
+ "learning_rate": 4.4583333333333336e-06,
270
+ "loss": 0.0006585324299521744,
271
+ "step": 14,
272
+ "step_time": 0.2181926339999336
273
+ },
274
+ {
275
+ "clip_ratio/high_max": 0.01888020895421505,
276
+ "clip_ratio/high_mean": 0.01888020895421505,
277
+ "clip_ratio/low_mean": 0.0,
278
+ "clip_ratio/low_min": 0.0,
279
+ "clip_ratio/region_mean": 0.01888020895421505,
280
+ "entropy": 4.414994716644287,
281
+ "epoch": 0.625,
282
+ "grad_norm": 1.1328541040420532,
283
+ "kl": 0.09972727298736572,
284
+ "learning_rate": 4.416666666666667e-06,
285
+ "loss": -0.0626036524772644,
286
+ "step": 15,
287
+ "step_time": 0.21817723499998465
288
+ },
289
+ {
290
+ "clip_ratio/high_max": 0.0,
291
+ "clip_ratio/high_mean": 0.0,
292
+ "clip_ratio/low_mean": 0.0,
293
+ "clip_ratio/low_min": 0.0,
294
+ "clip_ratio/region_mean": 0.0,
295
+ "entropy": 4.725884437561035,
296
+ "epoch": 0.6666666666666666,
297
+ "grad_norm": 0.02289302460849285,
298
+ "kl": 0.08768561482429504,
299
+ "learning_rate": 4.3750000000000005e-06,
300
+ "loss": 0.00021921405277680606,
301
+ "step": 16,
302
+ "step_time": 0.21883753800011618
303
+ },
304
+ {
305
+ "clip_ratio/high_max": 0.0,
306
+ "clip_ratio/high_mean": 0.0,
307
+ "clip_ratio/low_mean": 0.0,
308
+ "clip_ratio/low_min": 0.0,
309
+ "clip_ratio/region_mean": 0.0,
310
+ "completions/clipped_ratio": 1.0,
311
+ "completions/max_length": 384.0,
312
+ "completions/max_terminated_length": 0.0,
313
+ "completions/mean_length": 384.0,
314
+ "completions/mean_terminated_length": 0.0,
315
+ "completions/min_length": 384.0,
316
+ "completions/min_terminated_length": 0.0,
317
+ "entropy": 4.040867328643799,
318
+ "epoch": 0.7083333333333334,
319
+ "frac_reward_zero_std": 0.5,
320
+ "grad_norm": 2.39467191696167,
321
+ "kl": 0.1182798221707344,
322
+ "learning_rate": 4.333333333333334e-06,
323
+ "loss": 0.09359735995531082,
324
+ "num_tokens": 125920.0,
325
+ "reward": -0.05625000223517418,
326
+ "reward_std": 0.019364915788173676,
327
+ "rewards/GeneratorRewardFunction/mean": -0.05625000223517418,
328
+ "rewards/GeneratorRewardFunction/std": 0.019364915788173676,
329
+ "step": 17,
330
+ "step_time": 12.371318035000058
331
+ },
332
+ {
333
+ "clip_ratio/high_max": 0.0013020833721384406,
334
+ "clip_ratio/high_mean": 0.0013020833721384406,
335
+ "clip_ratio/low_mean": 0.0,
336
+ "clip_ratio/low_min": 0.0,
337
+ "clip_ratio/region_mean": 0.0013020833721384406,
338
+ "entropy": 4.358741760253906,
339
+ "epoch": 0.75,
340
+ "grad_norm": 1.4153165817260742,
341
+ "kl": 0.1291649043560028,
342
+ "learning_rate": 4.2916666666666665e-06,
343
+ "loss": -0.09299901872873306,
344
+ "step": 18,
345
+ "step_time": 0.220845523999742
346
+ },
347
+ {
348
+ "clip_ratio/high_max": 0.01171875,
349
+ "clip_ratio/high_mean": 0.01171875,
350
+ "clip_ratio/low_mean": 0.0,
351
+ "clip_ratio/low_min": 0.0,
352
+ "clip_ratio/region_mean": 0.01171875,
353
+ "entropy": 4.570446014404297,
354
+ "epoch": 0.7916666666666666,
355
+ "grad_norm": 1.1514215469360352,
356
+ "kl": 0.11547613888978958,
357
+ "learning_rate": 4.25e-06,
358
+ "loss": -0.061982929706573486,
359
+ "step": 19,
360
+ "step_time": 0.2208292430000256
361
+ },
362
+ {
363
+ "clip_ratio/high_max": 0.01627604104578495,
364
+ "clip_ratio/high_mean": 0.01627604104578495,
365
+ "clip_ratio/low_mean": 0.011067708022892475,
366
+ "clip_ratio/low_min": 0.011067708022892475,
367
+ "clip_ratio/region_mean": 0.02734375,
368
+ "entropy": 3.9513370990753174,
369
+ "epoch": 0.8333333333333334,
370
+ "grad_norm": 2.3550095558166504,
371
+ "kl": 0.1719018667936325,
372
+ "learning_rate": 4.208333333333333e-06,
373
+ "loss": 0.06247749552130699,
374
+ "step": 20,
375
+ "step_time": 0.22178658199982237
376
+ },
377
+ {
378
+ "clip_ratio/high_max": 0.0,
379
+ "clip_ratio/high_mean": 0.0,
380
+ "clip_ratio/low_mean": 0.0,
381
+ "clip_ratio/low_min": 0.0,
382
+ "clip_ratio/region_mean": 0.0,
383
+ "completions/clipped_ratio": 1.0,
384
+ "completions/max_length": 384.0,
385
+ "completions/max_terminated_length": 0.0,
386
+ "completions/mean_length": 384.0,
387
+ "completions/mean_terminated_length": 0.0,
388
+ "completions/min_length": 384.0,
389
+ "completions/min_terminated_length": 0.0,
390
+ "entropy": 3.850313901901245,
391
+ "epoch": 0.875,
392
+ "frac_reward_zero_std": 0.0,
393
+ "grad_norm": 2.7302346229553223,
394
+ "kl": 0.1612752377986908,
395
+ "learning_rate": 4.166666666666667e-06,
396
+ "loss": -0.10781875252723694,
397
+ "num_tokens": 150972.0,
398
+ "reward": -0.27531248331069946,
399
+ "reward_std": 0.29975247383117676,
400
+ "rewards/GeneratorRewardFunction/mean": -0.27531248331069946,
401
+ "rewards/GeneratorRewardFunction/std": 0.29975250363349915,
402
+ "step": 21,
403
+ "step_time": 12.298712889999933
404
+ },
405
+ {
406
+ "clip_ratio/high_max": 0.0052083334885537624,
407
+ "clip_ratio/high_mean": 0.0052083334885537624,
408
+ "clip_ratio/low_mean": 0.0,
409
+ "clip_ratio/low_min": 0.0,
410
+ "clip_ratio/region_mean": 0.0052083334885537624,
411
+ "entropy": 3.952542543411255,
412
+ "epoch": 0.9166666666666666,
413
+ "grad_norm": 2.175267457962036,
414
+ "kl": 0.15720854699611664,
415
+ "learning_rate": 4.125e-06,
416
+ "loss": -0.061483949422836304,
417
+ "step": 22,
418
+ "step_time": 0.2193120709998766
419
+ },
420
+ {
421
+ "clip_ratio/high_max": 0.0032552082557231188,
422
+ "clip_ratio/high_mean": 0.0032552082557231188,
423
+ "clip_ratio/low_mean": 0.009765625,
424
+ "clip_ratio/low_min": 0.009765625,
425
+ "clip_ratio/region_mean": 0.013020833022892475,
426
+ "entropy": 4.322933673858643,
427
+ "epoch": 0.9583333333333334,
428
+ "grad_norm": 3.3126373291015625,
429
+ "kl": 0.1569279283285141,
430
+ "learning_rate": 4.083333333333334e-06,
431
+ "loss": 0.17323081195354462,
432
+ "step": 23,
433
+ "step_time": 0.21882151099998737
434
+ },
435
+ {
436
+ "clip_ratio/high_max": 0.0572916679084301,
437
+ "clip_ratio/high_mean": 0.0572916679084301,
438
+ "clip_ratio/low_mean": 0.0078125,
439
+ "clip_ratio/low_min": 0.0078125,
440
+ "clip_ratio/region_mean": 0.0651041641831398,
441
+ "entropy": 4.215339660644531,
442
+ "epoch": 1.0,
443
+ "grad_norm": 3.0476796627044678,
444
+ "kl": 0.197212815284729,
445
+ "learning_rate": 4.041666666666667e-06,
446
+ "loss": 0.0013918876647949219,
447
+ "step": 24,
448
+ "step_time": 0.22078252000028442
449
+ },
450
+ {
451
+ "clip_ratio/high_max": 0.0,
452
+ "clip_ratio/high_mean": 0.0,
453
+ "clip_ratio/low_mean": 0.0,
454
+ "clip_ratio/low_min": 0.0,
455
+ "clip_ratio/region_mean": 0.0,
456
+ "completions/clipped_ratio": 1.0,
457
+ "completions/max_length": 384.0,
458
+ "completions/max_terminated_length": 0.0,
459
+ "completions/mean_length": 384.0,
460
+ "completions/mean_terminated_length": 0.0,
461
+ "completions/min_length": 384.0,
462
+ "completions/min_terminated_length": 0.0,
463
+ "entropy": 4.335132598876953,
464
+ "epoch": 1.0416666666666667,
465
+ "frac_reward_zero_std": 0.25,
466
+ "grad_norm": 2.326383113861084,
467
+ "kl": 0.16707448661327362,
468
+ "learning_rate": 4.000000000000001e-06,
469
+ "loss": 0.05758856609463692,
470
+ "num_tokens": 176436.0,
471
+ "reward": -0.12687499821186066,
472
+ "reward_std": 0.20430266857147217,
473
+ "rewards/GeneratorRewardFunction/mean": -0.12687499821186066,
474
+ "rewards/GeneratorRewardFunction/std": 0.20430268347263336,
475
+ "step": 25,
476
+ "step_time": 12.420606664000388
477
+ },
478
+ {
479
+ "clip_ratio/high_max": 0.0032552082557231188,
480
+ "clip_ratio/high_mean": 0.0032552082557231188,
481
+ "clip_ratio/low_mean": 0.0,
482
+ "clip_ratio/low_min": 0.0,
483
+ "clip_ratio/region_mean": 0.0032552082557231188,
484
+ "entropy": 4.311023712158203,
485
+ "epoch": 1.0833333333333333,
486
+ "grad_norm": 2.787942409515381,
487
+ "kl": 0.1736934781074524,
488
+ "learning_rate": 3.958333333333333e-06,
489
+ "loss": 0.0001516403426649049,
490
+ "step": 26,
491
+ "step_time": 0.2286363840003105
492
+ },
493
+ {
494
+ "clip_ratio/high_max": 0.009114583022892475,
495
+ "clip_ratio/high_mean": 0.009114583022892475,
496
+ "clip_ratio/low_mean": 0.0032552082557231188,
497
+ "clip_ratio/low_min": 0.0032552082557231188,
498
+ "clip_ratio/region_mean": 0.012369791977107525,
499
+ "entropy": 4.165957927703857,
500
+ "epoch": 1.125,
501
+ "grad_norm": 2.553452968597412,
502
+ "kl": 0.2077862024307251,
503
+ "learning_rate": 3.916666666666667e-06,
504
+ "loss": 0.035733964294195175,
505
+ "step": 27,
506
+ "step_time": 0.22833154999989347
507
+ },
508
+ {
509
+ "clip_ratio/high_max": 0.0358072929084301,
510
+ "clip_ratio/high_mean": 0.0358072929084301,
511
+ "clip_ratio/low_mean": 0.0,
512
+ "clip_ratio/low_min": 0.0,
513
+ "clip_ratio/region_mean": 0.0358072929084301,
514
+ "entropy": 4.27438497543335,
515
+ "epoch": 1.1666666666666667,
516
+ "grad_norm": 1.3039305210113525,
517
+ "kl": 0.1941693276166916,
518
+ "learning_rate": 3.875e-06,
519
+ "loss": -0.09091150015592575,
520
+ "step": 28,
521
+ "step_time": 0.22796160400048393
522
+ },
523
+ {
524
+ "clip_ratio/high_max": 0.0,
525
+ "clip_ratio/high_mean": 0.0,
526
+ "clip_ratio/low_mean": 0.0,
527
+ "clip_ratio/low_min": 0.0,
528
+ "clip_ratio/region_mean": 0.0,
529
+ "completions/clipped_ratio": 1.0,
530
+ "completions/max_length": 384.0,
531
+ "completions/max_terminated_length": 0.0,
532
+ "completions/mean_length": 384.0,
533
+ "completions/mean_terminated_length": 0.0,
534
+ "completions/min_length": 384.0,
535
+ "completions/min_terminated_length": 0.0,
536
+ "entropy": 4.181336879730225,
537
+ "epoch": 1.2083333333333333,
538
+ "frac_reward_zero_std": 0.75,
539
+ "grad_norm": 1.4635995626449585,
540
+ "kl": 0.17187662422657013,
541
+ "learning_rate": 3.833333333333334e-06,
542
+ "loss": 0.05418387055397034,
543
+ "num_tokens": 201524.0,
544
+ "reward": -0.05312500149011612,
545
+ "reward_std": 0.008539125323295593,
546
+ "rewards/GeneratorRewardFunction/mean": -0.05312500149011612,
547
+ "rewards/GeneratorRewardFunction/std": 0.008539126254618168,
548
+ "step": 29,
549
+ "step_time": 12.34016204099953
550
+ },
551
+ {
552
+ "clip_ratio/high_max": 0.0,
553
+ "clip_ratio/high_mean": 0.0,
554
+ "clip_ratio/low_mean": 0.0,
555
+ "clip_ratio/low_min": 0.0,
556
+ "clip_ratio/region_mean": 0.0,
557
+ "entropy": 4.342355251312256,
558
+ "epoch": 1.25,
559
+ "grad_norm": 1.3520798683166504,
560
+ "kl": 0.19560372829437256,
561
+ "learning_rate": 3.7916666666666666e-06,
562
+ "loss": -0.05310399457812309,
563
+ "step": 30,
564
+ "step_time": 0.22077936900041095
565
+ },
566
+ {
567
+ "clip_ratio/high_max": 0.0,
568
+ "clip_ratio/high_mean": 0.0,
569
+ "clip_ratio/low_mean": 0.0,
570
+ "clip_ratio/low_min": 0.0,
571
+ "clip_ratio/region_mean": 0.0,
572
+ "entropy": 4.002376079559326,
573
+ "epoch": 1.2916666666666667,
574
+ "grad_norm": 0.03668312355875969,
575
+ "kl": 0.18400262296199799,
576
+ "learning_rate": 3.7500000000000005e-06,
577
+ "loss": 0.0004600065294653177,
578
+ "step": 31,
579
+ "step_time": 0.23486327800037543
580
+ },
581
+ {
582
+ "clip_ratio/high_max": 0.0078125,
583
+ "clip_ratio/high_mean": 0.0078125,
584
+ "clip_ratio/low_mean": 0.00390625,
585
+ "clip_ratio/low_min": 0.00390625,
586
+ "clip_ratio/region_mean": 0.01171875,
587
+ "entropy": 4.291225433349609,
588
+ "epoch": 1.3333333333333333,
589
+ "grad_norm": 1.7566320896148682,
590
+ "kl": 0.17402441799640656,
591
+ "learning_rate": 3.708333333333334e-06,
592
+ "loss": 0.0007328187930397689,
593
+ "step": 32,
594
+ "step_time": 0.21991783900011797
595
+ },
596
+ {
597
+ "clip_ratio/high_max": 0.0,
598
+ "clip_ratio/high_mean": 0.0,
599
+ "clip_ratio/low_mean": 0.0,
600
+ "clip_ratio/low_min": 0.0,
601
+ "clip_ratio/region_mean": 0.0,
602
+ "completions/clipped_ratio": 1.0,
603
+ "completions/max_length": 384.0,
604
+ "completions/max_terminated_length": 0.0,
605
+ "completions/mean_length": 384.0,
606
+ "completions/mean_terminated_length": 0.0,
607
+ "completions/min_length": 384.0,
608
+ "completions/min_terminated_length": 0.0,
609
+ "entropy": 4.679595470428467,
610
+ "epoch": 1.375,
611
+ "frac_reward_zero_std": 1.0,
612
+ "grad_norm": 0.03962837904691696,
613
+ "kl": 0.18714486062526703,
614
+ "learning_rate": 3.6666666666666666e-06,
615
+ "loss": 0.00046786211896687746,
616
+ "num_tokens": 226532.0,
617
+ "reward": -0.05000000074505806,
618
+ "reward_std": 0.0,
619
+ "rewards/GeneratorRewardFunction/mean": -0.05000000074505806,
620
+ "rewards/GeneratorRewardFunction/std": 0.0,
621
+ "step": 33,
622
+ "step_time": 12.39098299699981
623
+ },
624
+ {
625
+ "clip_ratio/high_max": 0.0,
626
+ "clip_ratio/high_mean": 0.0,
627
+ "clip_ratio/low_mean": 0.0,
628
+ "clip_ratio/low_min": 0.0,
629
+ "clip_ratio/region_mean": 0.0,
630
+ "entropy": 4.106312274932861,
631
+ "epoch": 1.4166666666666667,
632
+ "grad_norm": 0.036486461758613586,
633
+ "kl": 0.1916690319776535,
634
+ "learning_rate": 3.625e-06,
635
+ "loss": 0.0004791725368704647,
636
+ "step": 34,
637
+ "step_time": 0.22225405700010015
638
+ },
639
+ {
640
+ "clip_ratio/high_max": 0.0,
641
+ "clip_ratio/high_mean": 0.0,
642
+ "clip_ratio/low_mean": 0.0,
643
+ "clip_ratio/low_min": 0.0,
644
+ "clip_ratio/region_mean": 0.0,
645
+ "entropy": 4.338494300842285,
646
+ "epoch": 1.4583333333333333,
647
+ "grad_norm": 0.04015596583485603,
648
+ "kl": 0.17945243418216705,
649
+ "learning_rate": 3.5833333333333335e-06,
650
+ "loss": 0.00044863109360449016,
651
+ "step": 35,
652
+ "step_time": 0.22212749400023313
653
+ },
654
+ {
655
+ "clip_ratio/high_max": 0.0,
656
+ "clip_ratio/high_mean": 0.0,
657
+ "clip_ratio/low_mean": 0.0,
658
+ "clip_ratio/low_min": 0.0,
659
+ "clip_ratio/region_mean": 0.0,
660
+ "entropy": 4.584677219390869,
661
+ "epoch": 1.5,
662
+ "grad_norm": 0.05036609247326851,
663
+ "kl": 0.21357925236225128,
664
+ "learning_rate": 3.5416666666666673e-06,
665
+ "loss": 0.0005339480703696609,
666
+ "step": 36,
667
+ "step_time": 0.22190836000027048
668
+ },
669
+ {
670
+ "clip_ratio/high_max": 0.0,
671
+ "clip_ratio/high_mean": 0.0,
672
+ "clip_ratio/low_mean": 0.0,
673
+ "clip_ratio/low_min": 0.0,
674
+ "clip_ratio/region_mean": 0.0,
675
+ "completions/clipped_ratio": 1.0,
676
+ "completions/max_length": 384.0,
677
+ "completions/max_terminated_length": 0.0,
678
+ "completions/mean_length": 384.0,
679
+ "completions/mean_terminated_length": 0.0,
680
+ "completions/min_length": 384.0,
681
+ "completions/min_terminated_length": 0.0,
682
+ "entropy": 4.206821918487549,
683
+ "epoch": 1.5416666666666665,
684
+ "frac_reward_zero_std": 0.75,
685
+ "grad_norm": 2.2474279403686523,
686
+ "kl": 0.17469431459903717,
687
+ "learning_rate": 3.5e-06,
688
+ "loss": 0.09434103965759277,
689
+ "num_tokens": 251600.0,
690
+ "reward": -0.08749999850988388,
691
+ "reward_std": 0.14999999105930328,
692
+ "rewards/GeneratorRewardFunction/mean": -0.08749999850988388,
693
+ "rewards/GeneratorRewardFunction/std": 0.15000000596046448,
694
+ "step": 37,
695
+ "step_time": 12.279682073000004
696
+ },
697
+ {
698
+ "clip_ratio/high_max": 0.0,
699
+ "clip_ratio/high_mean": 0.0,
700
+ "clip_ratio/low_mean": 0.0,
701
+ "clip_ratio/low_min": 0.0,
702
+ "clip_ratio/region_mean": 0.0,
703
+ "entropy": 4.365553379058838,
704
+ "epoch": 1.5833333333333335,
705
+ "grad_norm": 0.7859151363372803,
706
+ "kl": 0.20030474662780762,
707
+ "learning_rate": 3.4583333333333334e-06,
708
+ "loss": -0.03081577830016613,
709
+ "step": 38,
710
+ "step_time": 0.22089524599959987
711
+ },
712
+ {
713
+ "clip_ratio/high_max": 0.0013020833721384406,
714
+ "clip_ratio/high_mean": 0.0013020833721384406,
715
+ "clip_ratio/low_mean": 0.0,
716
+ "clip_ratio/low_min": 0.0,
717
+ "clip_ratio/region_mean": 0.0013020833721384406,
718
+ "entropy": 4.566871166229248,
719
+ "epoch": 1.625,
720
+ "grad_norm": 0.7892794013023376,
721
+ "kl": 0.23452065885066986,
722
+ "learning_rate": 3.416666666666667e-06,
723
+ "loss": -0.03072456084191799,
724
+ "step": 39,
725
+ "step_time": 0.22001067800010787
726
+ },
727
+ {
728
+ "clip_ratio/high_max": 0.0052083334885537624,
729
+ "clip_ratio/high_mean": 0.0052083334885537624,
730
+ "clip_ratio/low_mean": 0.0,
731
+ "clip_ratio/low_min": 0.0,
732
+ "clip_ratio/region_mean": 0.0052083334885537624,
733
+ "entropy": 4.517628192901611,
734
+ "epoch": 1.6666666666666665,
735
+ "grad_norm": 0.7610399127006531,
736
+ "kl": 0.22405748069286346,
737
+ "learning_rate": 3.3750000000000003e-06,
738
+ "loss": -0.030617548152804375,
739
+ "step": 40,
740
+ "step_time": 0.22107800999947358
741
+ },
742
+ {
743
+ "clip_ratio/high_max": 0.0,
744
+ "clip_ratio/high_mean": 0.0,
745
+ "clip_ratio/low_mean": 0.0,
746
+ "clip_ratio/low_min": 0.0,
747
+ "clip_ratio/region_mean": 0.0,
748
+ "completions/clipped_ratio": 1.0,
749
+ "completions/max_length": 384.0,
750
+ "completions/max_terminated_length": 0.0,
751
+ "completions/mean_length": 384.0,
752
+ "completions/mean_terminated_length": 0.0,
753
+ "completions/min_length": 384.0,
754
+ "completions/min_terminated_length": 0.0,
755
+ "entropy": 4.207302093505859,
756
+ "epoch": 1.7083333333333335,
757
+ "frac_reward_zero_std": 1.0,
758
+ "grad_norm": 0.03834126889705658,
759
+ "kl": 0.20183412730693817,
760
+ "learning_rate": 3.3333333333333333e-06,
761
+ "loss": 0.000504585332237184,
762
+ "num_tokens": 276896.0,
763
+ "reward": -0.05000000074505806,
764
+ "reward_std": 0.0,
765
+ "rewards/GeneratorRewardFunction/mean": -0.05000000074505806,
766
+ "rewards/GeneratorRewardFunction/std": 0.0,
767
+ "step": 41,
768
+ "step_time": 12.462519400999554
769
+ },
770
+ {
771
+ "clip_ratio/high_max": 0.0,
772
+ "clip_ratio/high_mean": 0.0,
773
+ "clip_ratio/low_mean": 0.0,
774
+ "clip_ratio/low_min": 0.0,
775
+ "clip_ratio/region_mean": 0.0,
776
+ "entropy": 4.1995134353637695,
777
+ "epoch": 1.75,
778
+ "grad_norm": 0.03475196659564972,
779
+ "kl": 0.20933139324188232,
780
+ "learning_rate": 3.2916666666666668e-06,
781
+ "loss": 0.0005233284318819642,
782
+ "step": 42,
783
+ "step_time": 0.2306131320001441
784
+ },
785
+ {
786
+ "clip_ratio/high_max": 0.0,
787
+ "clip_ratio/high_mean": 0.0,
788
+ "clip_ratio/low_mean": 0.0,
789
+ "clip_ratio/low_min": 0.0,
790
+ "clip_ratio/region_mean": 0.0,
791
+ "entropy": 4.859399318695068,
792
+ "epoch": 1.7916666666666665,
793
+ "grad_norm": 0.043436359614133835,
794
+ "kl": 0.21264620125293732,
795
+ "learning_rate": 3.2500000000000002e-06,
796
+ "loss": 0.0005316155147738755,
797
+ "step": 43,
798
+ "step_time": 0.22955826899942622
799
+ },
800
+ {
801
+ "clip_ratio/high_max": 0.0,
802
+ "clip_ratio/high_mean": 0.0,
803
+ "clip_ratio/low_mean": 0.0,
804
+ "clip_ratio/low_min": 0.0,
805
+ "clip_ratio/region_mean": 0.0,
806
+ "entropy": 4.589152812957764,
807
+ "epoch": 1.8333333333333335,
808
+ "grad_norm": 0.043015655130147934,
809
+ "kl": 0.22540146112442017,
810
+ "learning_rate": 3.2083333333333337e-06,
811
+ "loss": 0.000563503650482744,
812
+ "step": 44,
813
+ "step_time": 0.22972263199972076
814
+ },
815
+ {
816
+ "clip_ratio/high_max": 0.0,
817
+ "clip_ratio/high_mean": 0.0,
818
+ "clip_ratio/low_mean": 0.0,
819
+ "clip_ratio/low_min": 0.0,
820
+ "clip_ratio/region_mean": 0.0,
821
+ "completions/clipped_ratio": 1.0,
822
+ "completions/max_length": 384.0,
823
+ "completions/max_terminated_length": 0.0,
824
+ "completions/mean_length": 384.0,
825
+ "completions/mean_terminated_length": 0.0,
826
+ "completions/min_length": 384.0,
827
+ "completions/min_terminated_length": 0.0,
828
+ "entropy": 4.461331844329834,
829
+ "epoch": 1.875,
830
+ "frac_reward_zero_std": 0.5,
831
+ "grad_norm": 0.03550349920988083,
832
+ "kl": 0.1928916722536087,
833
+ "learning_rate": 3.1666666666666667e-06,
834
+ "loss": 0.0004822291957680136,
835
+ "num_tokens": 301944.0,
836
+ "reward": -0.125,
837
+ "reward_std": 0.20493900775909424,
838
+ "rewards/GeneratorRewardFunction/mean": -0.125,
839
+ "rewards/GeneratorRewardFunction/std": 0.20493900775909424,
840
+ "step": 45,
841
+ "step_time": 12.369378186999711
842
+ },
843
+ {
844
+ "clip_ratio/high_max": 0.0006510416860692203,
845
+ "clip_ratio/high_mean": 0.0006510416860692203,
846
+ "clip_ratio/low_mean": 0.0,
847
+ "clip_ratio/low_min": 0.0,
848
+ "clip_ratio/region_mean": 0.0006510416860692203,
849
+ "entropy": 4.596534252166748,
850
+ "epoch": 1.9166666666666665,
851
+ "grad_norm": 1.093070149421692,
852
+ "kl": 0.1923203319311142,
853
+ "learning_rate": 3.125e-06,
854
+ "loss": -0.062203943729400635,
855
+ "step": 46,
856
+ "step_time": 0.22156830799940508
857
+ },
858
+ {
859
+ "clip_ratio/high_max": 0.0026041667442768812,
860
+ "clip_ratio/high_mean": 0.0026041667442768812,
861
+ "clip_ratio/low_mean": 0.0,
862
+ "clip_ratio/low_min": 0.0,
863
+ "clip_ratio/region_mean": 0.0026041667442768812,
864
+ "entropy": 4.4759135246276855,
865
+ "epoch": 1.9583333333333335,
866
+ "grad_norm": 2.8484206199645996,
867
+ "kl": 0.22040079534053802,
868
+ "learning_rate": 3.0833333333333336e-06,
869
+ "loss": 0.0003803384315688163,
870
+ "step": 47,
871
+ "step_time": 0.22012663999976212
872
+ },
873
+ {
874
+ "clip_ratio/high_max": 0.001953125,
875
+ "clip_ratio/high_mean": 0.001953125,
876
+ "clip_ratio/low_mean": 0.0013020833721384406,
877
+ "clip_ratio/low_min": 0.0013020833721384406,
878
+ "clip_ratio/region_mean": 0.0032552082557231188,
879
+ "entropy": 4.502986907958984,
880
+ "epoch": 2.0,
881
+ "grad_norm": 2.476116418838501,
882
+ "kl": 0.17874129116535187,
883
+ "learning_rate": 3.0416666666666666e-06,
884
+ "loss": 0.06300241500139236,
885
+ "step": 48,
886
+ "step_time": 0.22058342899981653
887
+ },
888
+ {
889
+ "clip_ratio/high_max": 0.0,
890
+ "clip_ratio/high_mean": 0.0,
891
+ "clip_ratio/low_mean": 0.0,
892
+ "clip_ratio/low_min": 0.0,
893
+ "clip_ratio/region_mean": 0.0,
894
+ "completions/clipped_ratio": 1.0,
895
+ "completions/max_length": 384.0,
896
+ "completions/max_terminated_length": 0.0,
897
+ "completions/mean_length": 384.0,
898
+ "completions/mean_terminated_length": 0.0,
899
+ "completions/min_length": 384.0,
900
+ "completions/min_terminated_length": 0.0,
901
+ "entropy": 4.52464485168457,
902
+ "epoch": 2.0416666666666665,
903
+ "frac_reward_zero_std": 0.75,
904
+ "grad_norm": 1.1072802543640137,
905
+ "kl": 0.17327140271663666,
906
+ "learning_rate": 3e-06,
907
+ "loss": -0.0618751160800457,
908
+ "num_tokens": 327088.0,
909
+ "reward": -0.05406250059604645,
910
+ "reward_std": 0.016249997541308403,
911
+ "rewards/GeneratorRewardFunction/mean": -0.05406250059604645,
912
+ "rewards/GeneratorRewardFunction/std": 0.0162500012665987,
913
+ "step": 49,
914
+ "step_time": 12.398161551999692
915
+ },
916
+ {
917
+ "clip_ratio/high_max": 0.0,
918
+ "clip_ratio/high_mean": 0.0,
919
+ "clip_ratio/low_mean": 0.0,
920
+ "clip_ratio/low_min": 0.0,
921
+ "clip_ratio/region_mean": 0.0,
922
+ "entropy": 4.568341255187988,
923
+ "epoch": 2.0833333333333335,
924
+ "grad_norm": 0.03384983912110329,
925
+ "kl": 0.19154612720012665,
926
+ "learning_rate": 2.9583333333333335e-06,
927
+ "loss": 0.0004788653168361634,
928
+ "step": 50,
929
+ "step_time": 0.2228688120003426
930
+ },
931
+ {
932
+ "clip_ratio/high_max": 0.0026041667442768812,
933
+ "clip_ratio/high_mean": 0.0026041667442768812,
934
+ "clip_ratio/low_mean": 0.0,
935
+ "clip_ratio/low_min": 0.0,
936
+ "clip_ratio/region_mean": 0.0026041667442768812,
937
+ "entropy": 4.437350749969482,
938
+ "epoch": 2.125,
939
+ "grad_norm": 0.7250102758407593,
940
+ "kl": 0.1893668919801712,
941
+ "learning_rate": 2.916666666666667e-06,
942
+ "loss": -0.030638111755251884,
943
+ "step": 51,
944
+ "step_time": 0.2229313830002866
945
+ },
946
+ {
947
+ "clip_ratio/high_max": 0.0,
948
+ "clip_ratio/high_mean": 0.0,
949
+ "clip_ratio/low_mean": 0.0,
950
+ "clip_ratio/low_min": 0.0,
951
+ "clip_ratio/region_mean": 0.0,
952
+ "entropy": 4.27592134475708,
953
+ "epoch": 2.1666666666666665,
954
+ "grad_norm": 2.2188467979431152,
955
+ "kl": 0.17189562320709229,
956
+ "learning_rate": 2.875e-06,
957
+ "loss": 0.09424998611211777,
958
+ "step": 52,
959
+ "step_time": 0.2230674660004297
960
+ },
961
+ {
962
+ "clip_ratio/high_max": 0.0,
963
+ "clip_ratio/high_mean": 0.0,
964
+ "clip_ratio/low_mean": 0.0,
965
+ "clip_ratio/low_min": 0.0,
966
+ "clip_ratio/region_mean": 0.0,
967
+ "completions/clipped_ratio": 1.0,
968
+ "completions/max_length": 384.0,
969
+ "completions/max_terminated_length": 0.0,
970
+ "completions/mean_length": 384.0,
971
+ "completions/mean_terminated_length": 0.0,
972
+ "completions/min_length": 384.0,
973
+ "completions/min_terminated_length": 0.0,
974
+ "entropy": 4.4265666007995605,
975
+ "epoch": 2.2083333333333335,
976
+ "frac_reward_zero_std": 1.0,
977
+ "grad_norm": 0.038845159113407135,
978
+ "kl": 0.21073134243488312,
979
+ "learning_rate": 2.8333333333333335e-06,
980
+ "loss": 0.0005268284003250301,
981
+ "num_tokens": 352328.0,
982
+ "reward": -0.05000000074505806,
983
+ "reward_std": 0.0,
984
+ "rewards/GeneratorRewardFunction/mean": -0.05000000074505806,
985
+ "rewards/GeneratorRewardFunction/std": 0.0,
986
+ "step": 53,
987
+ "step_time": 12.3143936510005
988
+ },
989
+ {
990
+ "clip_ratio/high_max": 0.0,
991
+ "clip_ratio/high_mean": 0.0,
992
+ "clip_ratio/low_mean": 0.0,
993
+ "clip_ratio/low_min": 0.0,
994
+ "clip_ratio/region_mean": 0.0,
995
+ "entropy": 4.656918525695801,
996
+ "epoch": 2.25,
997
+ "grad_norm": 0.0484224371612072,
998
+ "kl": 0.2176627665758133,
999
+ "learning_rate": 2.791666666666667e-06,
1000
+ "loss": 0.0005441569373942912,
1001
+ "step": 54,
1002
+ "step_time": 0.22369184699982725
1003
+ },
1004
+ {
1005
+ "clip_ratio/high_max": 0.0,
1006
+ "clip_ratio/high_mean": 0.0,
1007
+ "clip_ratio/low_mean": 0.0,
1008
+ "clip_ratio/low_min": 0.0,
1009
+ "clip_ratio/region_mean": 0.0,
1010
+ "entropy": 4.2594475746154785,
1011
+ "epoch": 2.2916666666666665,
1012
+ "grad_norm": 0.03165994957089424,
1013
+ "kl": 0.19008274376392365,
1014
+ "learning_rate": 2.7500000000000004e-06,
1015
+ "loss": 0.0004752068780362606,
1016
+ "step": 55,
1017
+ "step_time": 0.22343963099956454
1018
+ },
1019
+ {
1020
+ "clip_ratio/high_max": 0.0,
1021
+ "clip_ratio/high_mean": 0.0,
1022
+ "clip_ratio/low_mean": 0.0,
1023
+ "clip_ratio/low_min": 0.0,
1024
+ "clip_ratio/region_mean": 0.0,
1025
+ "entropy": 4.303634166717529,
1026
+ "epoch": 2.3333333333333335,
1027
+ "grad_norm": 0.03170448914170265,
1028
+ "kl": 0.18705779314041138,
1029
+ "learning_rate": 2.7083333333333334e-06,
1030
+ "loss": 0.00046764445141889155,
1031
+ "step": 56,
1032
+ "step_time": 0.22311556299973745
1033
+ },
1034
+ {
1035
+ "clip_ratio/high_max": 0.0,
1036
+ "clip_ratio/high_mean": 0.0,
1037
+ "clip_ratio/low_mean": 0.0,
1038
+ "clip_ratio/low_min": 0.0,
1039
+ "clip_ratio/region_mean": 0.0,
1040
+ "completions/clipped_ratio": 1.0,
1041
+ "completions/max_length": 384.0,
1042
+ "completions/max_terminated_length": 0.0,
1043
+ "completions/mean_length": 384.0,
1044
+ "completions/mean_terminated_length": 0.0,
1045
+ "completions/min_length": 384.0,
1046
+ "completions/min_terminated_length": 0.0,
1047
+ "entropy": 4.496265888214111,
1048
+ "epoch": 2.375,
1049
+ "frac_reward_zero_std": 0.5,
1050
+ "grad_norm": 2.193422794342041,
1051
+ "kl": 0.22647613286972046,
1052
+ "learning_rate": 2.666666666666667e-06,
1053
+ "loss": 0.09425372630357742,
1054
+ "num_tokens": 377544.0,
1055
+ "reward": -0.0690625011920929,
1056
+ "reward_std": 0.07492704689502716,
1057
+ "rewards/GeneratorRewardFunction/mean": -0.0690625011920929,
1058
+ "rewards/GeneratorRewardFunction/std": 0.07492704689502716,
1059
+ "step": 57,
1060
+ "step_time": 12.497021862999645
1061
+ },
1062
+ {
1063
+ "clip_ratio/high_max": 0.0,
1064
+ "clip_ratio/high_mean": 0.0,
1065
+ "clip_ratio/low_mean": 0.0,
1066
+ "clip_ratio/low_min": 0.0,
1067
+ "clip_ratio/region_mean": 0.0,
1068
+ "entropy": 4.573244094848633,
1069
+ "epoch": 2.4166666666666665,
1070
+ "grad_norm": 1.3519999980926514,
1071
+ "kl": 0.21581268310546875,
1072
+ "learning_rate": 2.6250000000000003e-06,
1073
+ "loss": -0.09204903244972229,
1074
+ "step": 58,
1075
+ "step_time": 0.23012131300038163
1076
+ },
1077
+ {
1078
+ "clip_ratio/high_max": 0.001953125,
1079
+ "clip_ratio/high_mean": 0.001953125,
1080
+ "clip_ratio/low_mean": 0.0,
1081
+ "clip_ratio/low_min": 0.0,
1082
+ "clip_ratio/region_mean": 0.001953125,
1083
+ "entropy": 4.067762851715088,
1084
+ "epoch": 2.4583333333333335,
1085
+ "grad_norm": 2.2663047313690186,
1086
+ "kl": 0.19602923095226288,
1087
+ "learning_rate": 2.5833333333333337e-06,
1088
+ "loss": 0.05952334776520729,
1089
+ "step": 59,
1090
+ "step_time": 0.23155366399987543
1091
+ },
1092
+ {
1093
+ "clip_ratio/high_max": 0.008463541977107525,
1094
+ "clip_ratio/high_mean": 0.008463541977107525,
1095
+ "clip_ratio/low_mean": 0.0,
1096
+ "clip_ratio/low_min": 0.0,
1097
+ "clip_ratio/region_mean": 0.008463541977107525,
1098
+ "entropy": 4.1496710777282715,
1099
+ "epoch": 2.5,
1100
+ "grad_norm": 1.0887759923934937,
1101
+ "kl": 0.2114473581314087,
1102
+ "learning_rate": 2.5416666666666668e-06,
1103
+ "loss": -0.05951349437236786,
1104
+ "step": 60,
1105
+ "step_time": 0.22960056100055226
1106
+ }
1107
+ ],
1108
+ "logging_steps": 1,
1109
+ "max_steps": 120,
1110
+ "num_input_tokens_seen": 377544,
1111
+ "num_train_epochs": 5,
1112
+ "save_steps": 30,
1113
+ "stateful_callbacks": {
1114
+ "TrainerControl": {
1115
+ "args": {
1116
+ "should_epoch_stop": false,
1117
+ "should_evaluate": false,
1118
+ "should_log": false,
1119
+ "should_save": true,
1120
+ "should_training_stop": false
1121
+ },
1122
+ "attributes": {}
1123
+ }
1124
+ },
1125
+ "total_flos": 0.0,
1126
+ "train_batch_size": 4,
1127
+ "trial_name": null,
1128
+ "trial_params": null
1129
+ }
self_play_hf_l40s_full/round_004/generator_train/checkpoint-60/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b5dc67a9771849b225a6392756d4a1f5c89c7689c4ad63ef1c3f5f8879020ff
3
+ size 7249