Hugging Face's logo

Siddeshwar1625
/
osint-checkpoints

Safetensors
Model card Files
xet
 Community
Siddeshwar1625 commited on
Commit
d5e2521
·
verified ·
1 Parent(s): c21c732

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_006/generator_train/checkpoint-60/chat_template.jinja +54 -0
  3. self_play_hf_l40s_full/round_006/generator_train/checkpoint-60/config.json +57 -0
  4. self_play_hf_l40s_full/round_006/generator_train/checkpoint-60/generation_config.json +13 -0
  5. self_play_hf_l40s_full/round_006/generator_train/checkpoint-60/model.safetensors +3 -0
  6. self_play_hf_l40s_full/round_006/generator_train/checkpoint-60/optimizer.pt +3 -0
  7. self_play_hf_l40s_full/round_006/generator_train/checkpoint-60/rng_state.pth +3 -0
  8. self_play_hf_l40s_full/round_006/generator_train/checkpoint-60/scheduler.pt +3 -0
  9. self_play_hf_l40s_full/round_006/generator_train/checkpoint-60/tokenizer.json +3 -0
  10. self_play_hf_l40s_full/round_006/generator_train/checkpoint-60/tokenizer_config.json +32 -0
  11. self_play_hf_l40s_full/round_006/generator_train/checkpoint-60/trainer_state.json +1129 -0
  12. self_play_hf_l40s_full/round_006/generator_train/checkpoint-60/training_args.bin +3 -0
.gitattributes CHANGED
@@ -93,3 +93,4 @@ self_play_hf_l40s_full/round_005/answerer_train/checkpoint-90/tokenizer.json fil
93
  self_play_hf_l40s_full/round_005/answerer_train/checkpoint-120/tokenizer.json filter=lfs diff=lfs merge=lfs -text
94
  self_play_hf_l40s_full/round_005/answerer_train/final_model/tokenizer.json filter=lfs diff=lfs merge=lfs -text
95
  self_play_hf_l40s_full/round_006/generator_train/checkpoint-30/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
93
  self_play_hf_l40s_full/round_005/answerer_train/checkpoint-120/tokenizer.json filter=lfs diff=lfs merge=lfs -text
94
  self_play_hf_l40s_full/round_005/answerer_train/final_model/tokenizer.json filter=lfs diff=lfs merge=lfs -text
95
  self_play_hf_l40s_full/round_006/generator_train/checkpoint-30/tokenizer.json filter=lfs diff=lfs merge=lfs -text
96
+ self_play_hf_l40s_full/round_006/generator_train/checkpoint-60/tokenizer.json filter=lfs diff=lfs merge=lfs -text
self_play_hf_l40s_full/round_006/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_006/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_006/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_006/generator_train/checkpoint-60/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de0be85329eef98f866c604a833926033c42886c4a09f235e673fb237203b0c1
3
+ size 1976163472
self_play_hf_l40s_full/round_006/generator_train/checkpoint-60/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f05536b4e8c980f10d9d1dc036201b519b3c05da4371c830149868241c22790b
3
+ size 3952509771
self_play_hf_l40s_full/round_006/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_006/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_006/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_006/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_006/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": 4.419156551361084,
26
+ "epoch": 0.041666666666666664,
27
+ "frac_reward_zero_std": 1.0,
28
+ "grad_norm": 0.0010621912078931928,
29
+ "kl": 0.000388016109354794,
30
+ "learning_rate": 5e-06,
31
+ "loss": 9.70040218817303e-07,
32
+ "num_tokens": 25556.0,
33
+ "reward": -0.05000000074505806,
34
+ "reward_std": 0.0,
35
+ "rewards/GeneratorRewardFunction/mean": -0.05000000074505806,
36
+ "rewards/GeneratorRewardFunction/std": 0.0,
37
+ "step": 1,
38
+ "step_time": 12.240020418000313
39
+ },
40
+ {
41
+ "clip_ratio/high_max": 0.0,
42
+ "clip_ratio/high_mean": 0.0,
43
+ "clip_ratio/low_mean": 0.0,
44
+ "clip_ratio/low_min": 0.0,
45
+ "clip_ratio/region_mean": 0.0,
46
+ "entropy": 4.5731282234191895,
47
+ "epoch": 0.08333333333333333,
48
+ "grad_norm": 0.008937706239521503,
49
+ "kl": 0.00917684193700552,
50
+ "learning_rate": 4.958333333333334e-06,
51
+ "loss": 2.294210389663931e-05,
52
+ "step": 2,
53
+ "step_time": 0.22796191399993404
54
+ },
55
+ {
56
+ "clip_ratio/high_max": 0.0,
57
+ "clip_ratio/high_mean": 0.0,
58
+ "clip_ratio/low_mean": 0.0,
59
+ "clip_ratio/low_min": 0.0,
60
+ "clip_ratio/region_mean": 0.0,
61
+ "entropy": 4.546818256378174,
62
+ "epoch": 0.125,
63
+ "grad_norm": 0.035122279077768326,
64
+ "kl": 0.03370319679379463,
65
+ "learning_rate": 4.9166666666666665e-06,
66
+ "loss": 8.425798296229914e-05,
67
+ "step": 3,
68
+ "step_time": 0.2268119000000297
69
+ },
70
+ {
71
+ "clip_ratio/high_max": 0.0,
72
+ "clip_ratio/high_mean": 0.0,
73
+ "clip_ratio/low_mean": 0.0,
74
+ "clip_ratio/low_min": 0.0,
75
+ "clip_ratio/region_mean": 0.0,
76
+ "entropy": 4.314895153045654,
77
+ "epoch": 0.16666666666666666,
78
+ "grad_norm": 0.0398935042321682,
79
+ "kl": 0.030106937512755394,
80
+ "learning_rate": 4.875e-06,
81
+ "loss": 7.526733679696918e-05,
82
+ "step": 4,
83
+ "step_time": 0.22681275899958564
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": 4.582925319671631,
99
+ "epoch": 0.20833333333333334,
100
+ "frac_reward_zero_std": 1.0,
101
+ "grad_norm": 0.010530068539083004,
102
+ "kl": 0.014865164645016193,
103
+ "learning_rate": 4.833333333333333e-06,
104
+ "loss": 3.716291030286811e-05,
105
+ "num_tokens": 50700.0,
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": 12.283316627000204
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": 4.653713226318359,
120
+ "epoch": 0.25,
121
+ "grad_norm": 0.008336829952895641,
122
+ "kl": 0.01457882672548294,
123
+ "learning_rate": 4.791666666666668e-06,
124
+ "loss": 3.6447065212996677e-05,
125
+ "step": 6,
126
+ "step_time": 0.2217760690000432
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": 4.560446262359619,
135
+ "epoch": 0.2916666666666667,
136
+ "grad_norm": 0.0281889159232378,
137
+ "kl": 0.0411984808743,
138
+ "learning_rate": 4.75e-06,
139
+ "loss": 0.0001029962077154778,
140
+ "step": 7,
141
+ "step_time": 0.22122331500031578
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": 4.5829596519470215,
150
+ "epoch": 0.3333333333333333,
151
+ "grad_norm": 0.01816079579293728,
152
+ "kl": 0.02516261301934719,
153
+ "learning_rate": 4.708333333333334e-06,
154
+ "loss": 6.290652527241036e-05,
155
+ "step": 8,
156
+ "step_time": 0.22042209800019918
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": 4.369012355804443,
172
+ "epoch": 0.375,
173
+ "frac_reward_zero_std": 1.0,
174
+ "grad_norm": 0.008493429981172085,
175
+ "kl": 0.015684768557548523,
176
+ "learning_rate": 4.666666666666667e-06,
177
+ "loss": 3.921191455447115e-05,
178
+ "num_tokens": 75860.0,
179
+ "reward": -0.05000000074505806,
180
+ "reward_std": 0.0,
181
+ "rewards/GeneratorRewardFunction/mean": -0.05000000074505806,
182
+ "rewards/GeneratorRewardFunction/std": 0.0,
183
+ "step": 9,
184
+ "step_time": 12.394048645999646
185
+ },
186
+ {
187
+ "clip_ratio/high_max": 0.0,
188
+ "clip_ratio/high_mean": 0.0,
189
+ "clip_ratio/low_mean": 0.0,
190
+ "clip_ratio/low_min": 0.0,
191
+ "clip_ratio/region_mean": 0.0,
192
+ "entropy": 4.415802955627441,
193
+ "epoch": 0.4166666666666667,
194
+ "grad_norm": 0.011094732210040092,
195
+ "kl": 0.019924873486161232,
196
+ "learning_rate": 4.625000000000001e-06,
197
+ "loss": 4.9812180805020034e-05,
198
+ "step": 10,
199
+ "step_time": 0.22683883599984256
200
+ },
201
+ {
202
+ "clip_ratio/high_max": 0.0,
203
+ "clip_ratio/high_mean": 0.0,
204
+ "clip_ratio/low_mean": 0.0,
205
+ "clip_ratio/low_min": 0.0,
206
+ "clip_ratio/region_mean": 0.0,
207
+ "entropy": 4.466436862945557,
208
+ "epoch": 0.4583333333333333,
209
+ "grad_norm": 0.008779467083513737,
210
+ "kl": 0.017840370535850525,
211
+ "learning_rate": 4.583333333333333e-06,
212
+ "loss": 4.460093259694986e-05,
213
+ "step": 11,
214
+ "step_time": 0.2262637949997952
215
+ },
216
+ {
217
+ "clip_ratio/high_max": 0.0,
218
+ "clip_ratio/high_mean": 0.0,
219
+ "clip_ratio/low_mean": 0.0,
220
+ "clip_ratio/low_min": 0.0,
221
+ "clip_ratio/region_mean": 0.0,
222
+ "entropy": 4.496238708496094,
223
+ "epoch": 0.5,
224
+ "grad_norm": 0.012762094847857952,
225
+ "kl": 0.020208822563290596,
226
+ "learning_rate": 4.541666666666667e-06,
227
+ "loss": 5.052205597166903e-05,
228
+ "step": 12,
229
+ "step_time": 0.22641648600074404
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": 4.47562837600708,
245
+ "epoch": 0.5416666666666666,
246
+ "frac_reward_zero_std": 1.0,
247
+ "grad_norm": 0.01036064326763153,
248
+ "kl": 0.027347290888428688,
249
+ "learning_rate": 4.5e-06,
250
+ "loss": 6.836823740741238e-05,
251
+ "num_tokens": 100948.0,
252
+ "reward": -0.05000000074505806,
253
+ "reward_std": 0.0,
254
+ "rewards/GeneratorRewardFunction/mean": -0.05000000074505806,
255
+ "rewards/GeneratorRewardFunction/std": 0.0,
256
+ "step": 13,
257
+ "step_time": 12.410987389000184
258
+ },
259
+ {
260
+ "clip_ratio/high_max": 0.0,
261
+ "clip_ratio/high_mean": 0.0,
262
+ "clip_ratio/low_mean": 0.0,
263
+ "clip_ratio/low_min": 0.0,
264
+ "clip_ratio/region_mean": 0.0,
265
+ "entropy": 4.549487590789795,
266
+ "epoch": 0.5833333333333334,
267
+ "grad_norm": 0.00998639315366745,
268
+ "kl": 0.02104916237294674,
269
+ "learning_rate": 4.4583333333333336e-06,
270
+ "loss": 5.262290869723074e-05,
271
+ "step": 14,
272
+ "step_time": 0.2225908870004787
273
+ },
274
+ {
275
+ "clip_ratio/high_max": 0.0,
276
+ "clip_ratio/high_mean": 0.0,
277
+ "clip_ratio/low_mean": 0.0,
278
+ "clip_ratio/low_min": 0.0,
279
+ "clip_ratio/region_mean": 0.0,
280
+ "entropy": 4.591811656951904,
281
+ "epoch": 0.625,
282
+ "grad_norm": 0.011028505861759186,
283
+ "kl": 0.02356032468378544,
284
+ "learning_rate": 4.416666666666667e-06,
285
+ "loss": 5.8900812291540205e-05,
286
+ "step": 15,
287
+ "step_time": 0.22203837599954568
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": 4.564819812774658,
296
+ "epoch": 0.6666666666666666,
297
+ "grad_norm": 0.012791248969733715,
298
+ "kl": 0.02974030375480652,
299
+ "learning_rate": 4.3750000000000005e-06,
300
+ "loss": 7.435074803652242e-05,
301
+ "step": 16,
302
+ "step_time": 0.22162450700034242
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": 4.649733066558838,
318
+ "epoch": 0.7083333333333334,
319
+ "frac_reward_zero_std": 0.75,
320
+ "grad_norm": 0.008563967421650887,
321
+ "kl": 0.02316378615796566,
322
+ "learning_rate": 4.333333333333334e-06,
323
+ "loss": 5.7909463066607714e-05,
324
+ "num_tokens": 126536.0,
325
+ "reward": -0.05039062350988388,
326
+ "reward_std": 0.0015625000232830644,
327
+ "rewards/GeneratorRewardFunction/mean": -0.05039062350988388,
328
+ "rewards/GeneratorRewardFunction/std": 0.0015624994412064552,
329
+ "step": 17,
330
+ "step_time": 12.360184032000689
331
+ },
332
+ {
333
+ "clip_ratio/high_max": 0.0,
334
+ "clip_ratio/high_mean": 0.0,
335
+ "clip_ratio/low_mean": 0.0,
336
+ "clip_ratio/low_min": 0.0,
337
+ "clip_ratio/region_mean": 0.0,
338
+ "entropy": 4.570621013641357,
339
+ "epoch": 0.75,
340
+ "grad_norm": 2.3772716522216797,
341
+ "kl": 0.012507900595664978,
342
+ "learning_rate": 4.2916666666666665e-06,
343
+ "loss": 0.06100784242153168,
344
+ "step": 18,
345
+ "step_time": 0.22766595499979303
346
+ },
347
+ {
348
+ "clip_ratio/high_max": 0.0,
349
+ "clip_ratio/high_mean": 0.0,
350
+ "clip_ratio/low_mean": 0.0,
351
+ "clip_ratio/low_min": 0.0,
352
+ "clip_ratio/region_mean": 0.0,
353
+ "entropy": 4.49678373336792,
354
+ "epoch": 0.7916666666666666,
355
+ "grad_norm": 0.008817911148071289,
356
+ "kl": 0.014840642921626568,
357
+ "learning_rate": 4.25e-06,
358
+ "loss": 3.710160672198981e-05,
359
+ "step": 19,
360
+ "step_time": 0.22760423399995489
361
+ },
362
+ {
363
+ "clip_ratio/high_max": 0.01822916604578495,
364
+ "clip_ratio/high_mean": 0.01822916604578495,
365
+ "clip_ratio/low_mean": 0.0,
366
+ "clip_ratio/low_min": 0.0,
367
+ "clip_ratio/region_mean": 0.01822916604578495,
368
+ "entropy": 4.48759126663208,
369
+ "epoch": 0.8333333333333334,
370
+ "grad_norm": 1.0209619998931885,
371
+ "kl": 0.024702997878193855,
372
+ "learning_rate": 4.208333333333333e-06,
373
+ "loss": -0.060188356786966324,
374
+ "step": 20,
375
+ "step_time": 0.2271009630003391
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": 4.2363433837890625,
391
+ "epoch": 0.875,
392
+ "frac_reward_zero_std": 0.75,
393
+ "grad_norm": 0.014351247809827328,
394
+ "kl": 0.02771741896867752,
395
+ "learning_rate": 4.166666666666667e-06,
396
+ "loss": 6.929355004103854e-05,
397
+ "num_tokens": 151852.0,
398
+ "reward": -0.050312504172325134,
399
+ "reward_std": 0.0012500007869675756,
400
+ "rewards/GeneratorRewardFunction/mean": -0.050312504172325134,
401
+ "rewards/GeneratorRewardFunction/std": 0.001249999739229679,
402
+ "step": 21,
403
+ "step_time": 12.3216531449998
404
+ },
405
+ {
406
+ "clip_ratio/high_max": 0.0,
407
+ "clip_ratio/high_mean": 0.0,
408
+ "clip_ratio/low_mean": 0.0,
409
+ "clip_ratio/low_min": 0.0,
410
+ "clip_ratio/region_mean": 0.0,
411
+ "entropy": 4.444155216217041,
412
+ "epoch": 0.9166666666666666,
413
+ "grad_norm": 0.020809059962630272,
414
+ "kl": 0.04267991706728935,
415
+ "learning_rate": 4.125e-06,
416
+ "loss": 0.0001066997938323766,
417
+ "step": 22,
418
+ "step_time": 0.22639075000006414
419
+ },
420
+ {
421
+ "clip_ratio/high_max": 0.0,
422
+ "clip_ratio/high_mean": 0.0,
423
+ "clip_ratio/low_mean": 0.02018229104578495,
424
+ "clip_ratio/low_min": 0.02018229104578495,
425
+ "clip_ratio/region_mean": 0.02018229104578495,
426
+ "entropy": 4.4836273193359375,
427
+ "epoch": 0.9583333333333334,
428
+ "grad_norm": 2.161269426345825,
429
+ "kl": 0.06042298674583435,
430
+ "learning_rate": 4.083333333333334e-06,
431
+ "loss": 0.09100092202425003,
432
+ "step": 23,
433
+ "step_time": 0.22619211599976552
434
+ },
435
+ {
436
+ "clip_ratio/high_max": 0.0579427070915699,
437
+ "clip_ratio/high_mean": 0.0579427070915699,
438
+ "clip_ratio/low_mean": 0.0,
439
+ "clip_ratio/low_min": 0.0,
440
+ "clip_ratio/region_mean": 0.0579427070915699,
441
+ "entropy": 4.380900859832764,
442
+ "epoch": 1.0,
443
+ "grad_norm": 1.2373660802841187,
444
+ "kl": 0.056122809648513794,
445
+ "learning_rate": 4.041666666666667e-06,
446
+ "loss": -0.08933254331350327,
447
+ "step": 24,
448
+ "step_time": 0.226315008999336
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": 4.485828399658203,
464
+ "epoch": 1.0416666666666667,
465
+ "frac_reward_zero_std": 1.0,
466
+ "grad_norm": 0.023255605250597,
467
+ "kl": 0.058413077145814896,
468
+ "learning_rate": 4.000000000000001e-06,
469
+ "loss": 0.0001460327039239928,
470
+ "num_tokens": 177116.0,
471
+ "reward": -0.05000000074505806,
472
+ "reward_std": 0.0,
473
+ "rewards/GeneratorRewardFunction/mean": -0.05000000074505806,
474
+ "rewards/GeneratorRewardFunction/std": 0.0,
475
+ "step": 25,
476
+ "step_time": 12.349761277000653
477
+ },
478
+ {
479
+ "clip_ratio/high_max": 0.0,
480
+ "clip_ratio/high_mean": 0.0,
481
+ "clip_ratio/low_mean": 0.0,
482
+ "clip_ratio/low_min": 0.0,
483
+ "clip_ratio/region_mean": 0.0,
484
+ "entropy": 4.218708515167236,
485
+ "epoch": 1.0833333333333333,
486
+ "grad_norm": 0.016422593966126442,
487
+ "kl": 0.04855342209339142,
488
+ "learning_rate": 3.958333333333333e-06,
489
+ "loss": 0.00012138354941271245,
490
+ "step": 26,
491
+ "step_time": 0.22763793499962048
492
+ },
493
+ {
494
+ "clip_ratio/high_max": 0.0,
495
+ "clip_ratio/high_mean": 0.0,
496
+ "clip_ratio/low_mean": 0.0,
497
+ "clip_ratio/low_min": 0.0,
498
+ "clip_ratio/region_mean": 0.0,
499
+ "entropy": 4.261697292327881,
500
+ "epoch": 1.125,
501
+ "grad_norm": 0.01774059794843197,
502
+ "kl": 0.056816667318344116,
503
+ "learning_rate": 3.916666666666667e-06,
504
+ "loss": 0.00014204165199771523,
505
+ "step": 27,
506
+ "step_time": 0.22682179799994628
507
+ },
508
+ {
509
+ "clip_ratio/high_max": 0.0,
510
+ "clip_ratio/high_mean": 0.0,
511
+ "clip_ratio/low_mean": 0.0,
512
+ "clip_ratio/low_min": 0.0,
513
+ "clip_ratio/region_mean": 0.0,
514
+ "entropy": 4.387914180755615,
515
+ "epoch": 1.1666666666666667,
516
+ "grad_norm": 0.018558742478489876,
517
+ "kl": 0.0531492717564106,
518
+ "learning_rate": 3.875e-06,
519
+ "loss": 0.00013287317415233701,
520
+ "step": 28,
521
+ "step_time": 0.22764357600044605
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": 4.574448585510254,
537
+ "epoch": 1.2083333333333333,
538
+ "frac_reward_zero_std": 1.0,
539
+ "grad_norm": 0.023394504562020302,
540
+ "kl": 0.06513897329568863,
541
+ "learning_rate": 3.833333333333334e-06,
542
+ "loss": 0.00016284745652228594,
543
+ "num_tokens": 202428.0,
544
+ "reward": -0.05000000074505806,
545
+ "reward_std": 0.0,
546
+ "rewards/GeneratorRewardFunction/mean": -0.05000000074505806,
547
+ "rewards/GeneratorRewardFunction/std": 0.0,
548
+ "step": 29,
549
+ "step_time": 12.334460025000226
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": 4.121253490447998,
558
+ "epoch": 1.25,
559
+ "grad_norm": 0.019013958051800728,
560
+ "kl": 0.07019994407892227,
561
+ "learning_rate": 3.7916666666666666e-06,
562
+ "loss": 0.00017549986660014838,
563
+ "step": 30,
564
+ "step_time": 0.22793087099944387
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": 4.393718719482422,
573
+ "epoch": 1.2916666666666667,
574
+ "grad_norm": 0.020440492779016495,
575
+ "kl": 0.06539925187826157,
576
+ "learning_rate": 3.7500000000000005e-06,
577
+ "loss": 0.00016349814541172236,
578
+ "step": 31,
579
+ "step_time": 0.2279218960002254
580
+ },
581
+ {
582
+ "clip_ratio/high_max": 0.0,
583
+ "clip_ratio/high_mean": 0.0,
584
+ "clip_ratio/low_mean": 0.0,
585
+ "clip_ratio/low_min": 0.0,
586
+ "clip_ratio/region_mean": 0.0,
587
+ "entropy": 4.2539963722229,
588
+ "epoch": 1.3333333333333333,
589
+ "grad_norm": 0.02275467850267887,
590
+ "kl": 0.08661907911300659,
591
+ "learning_rate": 3.708333333333334e-06,
592
+ "loss": 0.00021654769079759717,
593
+ "step": 32,
594
+ "step_time": 0.22655504700014717
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": 4.461463928222656,
610
+ "epoch": 1.375,
611
+ "frac_reward_zero_std": 1.0,
612
+ "grad_norm": 0.021151196211576462,
613
+ "kl": 0.08688134700059891,
614
+ "learning_rate": 3.6666666666666666e-06,
615
+ "loss": 0.0002172033564420417,
616
+ "num_tokens": 227684.0,
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": 12.324289608999607
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": 4.52423620223999,
631
+ "epoch": 1.4166666666666667,
632
+ "grad_norm": 0.025681816041469574,
633
+ "kl": 0.08941265195608139,
634
+ "learning_rate": 3.625e-06,
635
+ "loss": 0.0002235316060250625,
636
+ "step": 34,
637
+ "step_time": 0.22279499900014343
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": 4.341139316558838,
646
+ "epoch": 1.4583333333333333,
647
+ "grad_norm": 0.023534711450338364,
648
+ "kl": 0.0802784413099289,
649
+ "learning_rate": 3.5833333333333335e-06,
650
+ "loss": 0.0002006961003644392,
651
+ "step": 35,
652
+ "step_time": 0.22278804800043872
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": 4.264102458953857,
661
+ "epoch": 1.5,
662
+ "grad_norm": 0.021178873255848885,
663
+ "kl": 0.07734989374876022,
664
+ "learning_rate": 3.5416666666666673e-06,
665
+ "loss": 0.0001933747116709128,
666
+ "step": 36,
667
+ "step_time": 0.22261148399957165
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": 4.503022193908691,
683
+ "epoch": 1.5416666666666665,
684
+ "frac_reward_zero_std": 0.75,
685
+ "grad_norm": 2.258732557296753,
686
+ "kl": 0.11438712477684021,
687
+ "learning_rate": 3.5e-06,
688
+ "loss": 0.09415566176176071,
689
+ "num_tokens": 253224.0,
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": 12.366807610999786
696
+ },
697
+ {
698
+ "clip_ratio/high_max": 0.0006510416860692203,
699
+ "clip_ratio/high_mean": 0.0006510416860692203,
700
+ "clip_ratio/low_mean": 0.0,
701
+ "clip_ratio/low_min": 0.0,
702
+ "clip_ratio/region_mean": 0.0006510416860692203,
703
+ "entropy": 4.676230430603027,
704
+ "epoch": 1.5833333333333335,
705
+ "grad_norm": 0.7382159233093262,
706
+ "kl": 0.10603597015142441,
707
+ "learning_rate": 3.4583333333333334e-06,
708
+ "loss": -0.031013809144496918,
709
+ "step": 38,
710
+ "step_time": 0.22748129699994024
711
+ },
712
+ {
713
+ "clip_ratio/high_max": 0.0052083334885537624,
714
+ "clip_ratio/high_mean": 0.0052083334885537624,
715
+ "clip_ratio/low_mean": 0.0,
716
+ "clip_ratio/low_min": 0.0,
717
+ "clip_ratio/region_mean": 0.0052083334885537624,
718
+ "entropy": 4.558745861053467,
719
+ "epoch": 1.625,
720
+ "grad_norm": 0.7514893412590027,
721
+ "kl": 0.10781397670507431,
722
+ "learning_rate": 3.416666666666667e-06,
723
+ "loss": -0.03084901161491871,
724
+ "step": 39,
725
+ "step_time": 0.22712060100002418
726
+ },
727
+ {
728
+ "clip_ratio/high_max": 0.014322916977107525,
729
+ "clip_ratio/high_mean": 0.014322916977107525,
730
+ "clip_ratio/low_mean": 0.0,
731
+ "clip_ratio/low_min": 0.0,
732
+ "clip_ratio/region_mean": 0.014322916977107525,
733
+ "entropy": 4.700138092041016,
734
+ "epoch": 1.6666666666666665,
735
+ "grad_norm": 0.7201879620552063,
736
+ "kl": 0.10677776485681534,
737
+ "learning_rate": 3.3750000000000003e-06,
738
+ "loss": -0.030954331159591675,
739
+ "step": 40,
740
+ "step_time": 0.22728954300055193
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": 4.267068862915039,
756
+ "epoch": 1.7083333333333335,
757
+ "frac_reward_zero_std": 1.0,
758
+ "grad_norm": 0.0244586244225502,
759
+ "kl": 0.09562627226114273,
760
+ "learning_rate": 3.3333333333333333e-06,
761
+ "loss": 0.0002390656591160223,
762
+ "num_tokens": 278568.0,
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": 12.331937089000348
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": 4.210024356842041,
777
+ "epoch": 1.75,
778
+ "grad_norm": 0.02442288212478161,
779
+ "kl": 0.09661825746297836,
780
+ "learning_rate": 3.2916666666666668e-06,
781
+ "loss": 0.00024154565471690148,
782
+ "step": 42,
783
+ "step_time": 0.22924090299966338
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": 4.3304123878479,
792
+ "epoch": 1.7916666666666665,
793
+ "grad_norm": 0.025623051449656487,
794
+ "kl": 0.10711003094911575,
795
+ "learning_rate": 3.2500000000000002e-06,
796
+ "loss": 0.0002677750599104911,
797
+ "step": 43,
798
+ "step_time": 0.2279044349998003
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": 4.030674457550049,
807
+ "epoch": 1.8333333333333335,
808
+ "grad_norm": 0.02603721246123314,
809
+ "kl": 0.0915822982788086,
810
+ "learning_rate": 3.2083333333333337e-06,
811
+ "loss": 0.0002289557596668601,
812
+ "step": 44,
813
+ "step_time": 0.22760460999961651
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": 4.380483150482178,
829
+ "epoch": 1.875,
830
+ "frac_reward_zero_std": 1.0,
831
+ "grad_norm": 0.02638942189514637,
832
+ "kl": 0.10838159918785095,
833
+ "learning_rate": 3.1666666666666667e-06,
834
+ "loss": 0.0002709540131036192,
835
+ "num_tokens": 303704.0,
836
+ "reward": -0.05000000074505806,
837
+ "reward_std": 0.0,
838
+ "rewards/GeneratorRewardFunction/mean": -0.05000000074505806,
839
+ "rewards/GeneratorRewardFunction/std": 0.0,
840
+ "step": 45,
841
+ "step_time": 12.200919003000308
842
+ },
843
+ {
844
+ "clip_ratio/high_max": 0.0,
845
+ "clip_ratio/high_mean": 0.0,
846
+ "clip_ratio/low_mean": 0.0,
847
+ "clip_ratio/low_min": 0.0,
848
+ "clip_ratio/region_mean": 0.0,
849
+ "entropy": 4.636005878448486,
850
+ "epoch": 1.9166666666666665,
851
+ "grad_norm": 0.02821226790547371,
852
+ "kl": 0.11671914905309677,
853
+ "learning_rate": 3.125e-06,
854
+ "loss": 0.00029179788543842733,
855
+ "step": 46,
856
+ "step_time": 0.22051493800063326
857
+ },
858
+ {
859
+ "clip_ratio/high_max": 0.0,
860
+ "clip_ratio/high_mean": 0.0,
861
+ "clip_ratio/low_mean": 0.0,
862
+ "clip_ratio/low_min": 0.0,
863
+ "clip_ratio/region_mean": 0.0,
864
+ "entropy": 4.353121280670166,
865
+ "epoch": 1.9583333333333335,
866
+ "grad_norm": 0.025765936821699142,
867
+ "kl": 0.10733228921890259,
868
+ "learning_rate": 3.0833333333333336e-06,
869
+ "loss": 0.00026833071024157107,
870
+ "step": 47,
871
+ "step_time": 0.22105524000016885
872
+ },
873
+ {
874
+ "clip_ratio/high_max": 0.0,
875
+ "clip_ratio/high_mean": 0.0,
876
+ "clip_ratio/low_mean": 0.0,
877
+ "clip_ratio/low_min": 0.0,
878
+ "clip_ratio/region_mean": 0.0,
879
+ "entropy": 4.313055515289307,
880
+ "epoch": 2.0,
881
+ "grad_norm": 0.02216164767742157,
882
+ "kl": 0.11219236999750137,
883
+ "learning_rate": 3.0416666666666666e-06,
884
+ "loss": 0.00028048091917298734,
885
+ "step": 48,
886
+ "step_time": 0.22110724199956167
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": 4.3338704109191895,
902
+ "epoch": 2.0416666666666665,
903
+ "frac_reward_zero_std": 0.75,
904
+ "grad_norm": 0.02499038726091385,
905
+ "kl": 0.11569738388061523,
906
+ "learning_rate": 3e-06,
907
+ "loss": 0.00028924347134307027,
908
+ "num_tokens": 328856.0,
909
+ "reward": -0.08749999850988388,
910
+ "reward_std": 0.15000000596046448,
911
+ "rewards/GeneratorRewardFunction/mean": -0.08749999850988388,
912
+ "rewards/GeneratorRewardFunction/std": 0.15000000596046448,
913
+ "step": 49,
914
+ "step_time": 12.343941664999875
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": 4.5195746421813965,
923
+ "epoch": 2.0833333333333335,
924
+ "grad_norm": 2.4670119285583496,
925
+ "kl": 0.1422944813966751,
926
+ "learning_rate": 2.9583333333333335e-06,
927
+ "loss": 0.06323760002851486,
928
+ "step": 50,
929
+ "step_time": 0.22882165300052293
930
+ },
931
+ {
932
+ "clip_ratio/high_max": 0.0,
933
+ "clip_ratio/high_mean": 0.0,
934
+ "clip_ratio/low_mean": 0.0,
935
+ "clip_ratio/low_min": 0.0,
936
+ "clip_ratio/region_mean": 0.0,
937
+ "entropy": 4.506677150726318,
938
+ "epoch": 2.125,
939
+ "grad_norm": 1.0566953420639038,
940
+ "kl": 0.11809781938791275,
941
+ "learning_rate": 2.916666666666667e-06,
942
+ "loss": -0.061977166682481766,
943
+ "step": 51,
944
+ "step_time": 0.22738053699958982
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": 4.5825653076171875,
953
+ "epoch": 2.1666666666666665,
954
+ "grad_norm": 0.025074757635593414,
955
+ "kl": 0.11249041557312012,
956
+ "learning_rate": 2.875e-06,
957
+ "loss": 0.00028122603544034064,
958
+ "step": 52,
959
+ "step_time": 0.22735149699929025
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": 4.527835845947266,
975
+ "epoch": 2.2083333333333335,
976
+ "frac_reward_zero_std": 0.5,
977
+ "grad_norm": 2.6713056564331055,
978
+ "kl": 0.12111381441354752,
979
+ "learning_rate": 2.8333333333333335e-06,
980
+ "loss": 0.03154236823320389,
981
+ "num_tokens": 353996.0,
982
+ "reward": -0.125,
983
+ "reward_std": 0.20493900775909424,
984
+ "rewards/GeneratorRewardFunction/mean": -0.125,
985
+ "rewards/GeneratorRewardFunction/std": 0.20493900775909424,
986
+ "step": 53,
987
+ "step_time": 12.368164391999926
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": 4.436413288116455,
996
+ "epoch": 2.25,
997
+ "grad_norm": 2.4797263145446777,
998
+ "kl": 0.12897814810276031,
999
+ "learning_rate": 2.791666666666667e-06,
1000
+ "loss": 0.03157377243041992,
1001
+ "step": 54,
1002
+ "step_time": 0.2241763499996523
1003
+ },
1004
+ {
1005
+ "clip_ratio/high_max": 0.0045572915114462376,
1006
+ "clip_ratio/high_mean": 0.0045572915114462376,
1007
+ "clip_ratio/low_mean": 0.0,
1008
+ "clip_ratio/low_min": 0.0,
1009
+ "clip_ratio/region_mean": 0.0045572915114462376,
1010
+ "entropy": 4.4373016357421875,
1011
+ "epoch": 2.2916666666666665,
1012
+ "grad_norm": 0.7626734972000122,
1013
+ "kl": 0.12167340517044067,
1014
+ "learning_rate": 2.7500000000000004e-06,
1015
+ "loss": -0.030798563733696938,
1016
+ "step": 55,
1017
+ "step_time": 0.22353983800076094
1018
+ },
1019
+ {
1020
+ "clip_ratio/high_max": 0.008463541977107525,
1021
+ "clip_ratio/high_mean": 0.008463541977107525,
1022
+ "clip_ratio/low_mean": 0.0,
1023
+ "clip_ratio/low_min": 0.0,
1024
+ "clip_ratio/region_mean": 0.008463541977107525,
1025
+ "entropy": 4.537744045257568,
1026
+ "epoch": 2.3333333333333335,
1027
+ "grad_norm": 0.759864091873169,
1028
+ "kl": 0.12279677391052246,
1029
+ "learning_rate": 2.7083333333333334e-06,
1030
+ "loss": -0.03098411113023758,
1031
+ "step": 56,
1032
+ "step_time": 0.22335842400025285
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": 4.574461460113525,
1048
+ "epoch": 2.375,
1049
+ "frac_reward_zero_std": 1.0,
1050
+ "grad_norm": 0.02886245585978031,
1051
+ "kl": 0.13087470829486847,
1052
+ "learning_rate": 2.666666666666667e-06,
1053
+ "loss": 0.0003271868044976145,
1054
+ "num_tokens": 379372.0,
1055
+ "reward": -0.05000000074505806,
1056
+ "reward_std": 0.0,
1057
+ "rewards/GeneratorRewardFunction/mean": -0.05000000074505806,
1058
+ "rewards/GeneratorRewardFunction/std": 0.0,
1059
+ "step": 57,
1060
+ "step_time": 12.374098764999871
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": 4.476459980010986,
1069
+ "epoch": 2.4166666666666665,
1070
+ "grad_norm": 0.022171108052134514,
1071
+ "kl": 0.10779599100351334,
1072
+ "learning_rate": 2.6250000000000003e-06,
1073
+ "loss": 0.00026949000312015414,
1074
+ "step": 58,
1075
+ "step_time": 0.22852272799991624
1076
+ },
1077
+ {
1078
+ "clip_ratio/high_max": 0.0,
1079
+ "clip_ratio/high_mean": 0.0,
1080
+ "clip_ratio/low_mean": 0.0,
1081
+ "clip_ratio/low_min": 0.0,
1082
+ "clip_ratio/region_mean": 0.0,
1083
+ "entropy": 4.3449883460998535,
1084
+ "epoch": 2.4583333333333335,
1085
+ "grad_norm": 0.02181490883231163,
1086
+ "kl": 0.11665940284729004,
1087
+ "learning_rate": 2.5833333333333337e-06,
1088
+ "loss": 0.0002916484954766929,
1089
+ "step": 59,
1090
+ "step_time": 0.22875684299924615
1091
+ },
1092
+ {
1093
+ "clip_ratio/high_max": 0.0,
1094
+ "clip_ratio/high_mean": 0.0,
1095
+ "clip_ratio/low_mean": 0.0,
1096
+ "clip_ratio/low_min": 0.0,
1097
+ "clip_ratio/region_mean": 0.0,
1098
+ "entropy": 4.47904109954834,
1099
+ "epoch": 2.5,
1100
+ "grad_norm": 0.02400827780365944,
1101
+ "kl": 0.1155659556388855,
1102
+ "learning_rate": 2.5416666666666668e-06,
1103
+ "loss": 0.00028891488909721375,
1104
+ "step": 60,
1105
+ "step_time": 0.22864042999935918
1106
+ }
1107
+ ],
1108
+ "logging_steps": 1,
1109
+ "max_steps": 120,
1110
+ "num_input_tokens_seen": 379372,
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_006/generator_train/checkpoint-60/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62dc5288cb3b8467d15382d666c4c904651b55d91aa621c4ccdf315cb995506a
3
+ size 7249