thuannh commited on
Commit
d1c007a
·
verified ·
1 Parent(s): d50a8a5

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
added_tokens.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 151668,
3
+ "</tool_call>": 151658,
4
+ "</tool_response>": 151666,
5
+ "<think>": 151667,
6
+ "<tool_call>": 151657,
7
+ "<tool_response>": 151665,
8
+ "<|box_end|>": 151649,
9
+ "<|box_start|>": 151648,
10
+ "<|endoftext|>": 151643,
11
+ "<|file_sep|>": 151664,
12
+ "<|fim_middle|>": 151660,
13
+ "<|fim_pad|>": 151662,
14
+ "<|fim_prefix|>": 151659,
15
+ "<|fim_suffix|>": 151661,
16
+ "<|im_end|>": 151645,
17
+ "<|im_start|>": 151644,
18
+ "<|image_pad|>": 151655,
19
+ "<|object_ref_end|>": 151647,
20
+ "<|object_ref_start|>": 151646,
21
+ "<|quad_end|>": 151651,
22
+ "<|quad_start|>": 151650,
23
+ "<|repo_name|>": 151663,
24
+ "<|video_pad|>": 151656,
25
+ "<|vision_end|>": 151653,
26
+ "<|vision_pad|>": 151654,
27
+ "<|vision_start|>": 151652
28
+ }
chat_template.jinja ADDED
@@ -0,0 +1,100 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0].role == 'system' %}
4
+ {{- messages[0].content + '\n\n' }}
5
+ {%- endif %}
6
+ {{- "# 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>" }}
7
+ {%- for tool in tools %}
8
+ {{- "\n" }}
9
+ {{- tool | tojson }}
10
+ {%- endfor %}
11
+ {{- "\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" }}
12
+ {%- else %}
13
+ {%- if messages[0].role == 'system' %}
14
+ {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
15
+ {%- endif %}
16
+ {%- endif %}
17
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
18
+ {%- for message in messages[::-1] %}
19
+ {%- set index = (messages|length - 1) - loop.index0 %}
20
+ {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
21
+ {%- set ns.multi_step_tool = false %}
22
+ {%- set ns.last_query_index = index %}
23
+ {%- endif %}
24
+ {%- endfor %}
25
+ {%- for message in messages %}
26
+ {%- if message.content is string %}
27
+ {%- set content = message.content %}
28
+ {%- else %}
29
+ {%- set content = '' %}
30
+ {%- endif %}
31
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
32
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
33
+ {%- elif message.role == "assistant" %}
34
+ {%- set reasoning_content = '' %}
35
+ {%- if message.reasoning_content is string %}
36
+ {%- set reasoning_content = message.reasoning_content %}
37
+ {%- else %}
38
+ {%- if '</think>' in content %}
39
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
40
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
41
+ {%- endif %}
42
+ {%- endif %}
43
+ {%- if loop.index0 > ns.last_query_index %}
44
+ {%- if loop.last or (not loop.last and reasoning_content) %}
45
+ {{- '<|im_start|>' + message.role }}
46
+ {% generation %}
47
+ {{- '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
48
+ {% endgeneration %}
49
+ {%- else %}
50
+ {{- '<|im_start|>' + message.role + '\n' }}
51
+ {% generation %}
52
+ {{- content }}
53
+ {% endgeneration %}
54
+ {%- endif %}
55
+ {%- else %}
56
+ {{- '<|im_start|>' + message.role + '\n' }}
57
+ {% generation %}
58
+ {{- content }}
59
+ {% endgeneration %}
60
+ {%- endif %}
61
+ {% generation %}
62
+ {%- if message.tool_calls %}
63
+ {%- for tool_call in message.tool_calls %}
64
+ {%- if (loop.first and content) or (not loop.first) %}
65
+ {{- '\n' }}
66
+ {%- endif %}
67
+ {%- if tool_call.function %}
68
+ {%- set tool_call = tool_call.function %}
69
+ {%- endif %}
70
+ {{- '<tool_call>\n{"name": "' }}
71
+ {{- tool_call.name }}
72
+ {{- '", "arguments": ' }}
73
+ {%- if tool_call.arguments is string %}
74
+ {{- tool_call.arguments }}
75
+ {%- else %}
76
+ {{- tool_call.arguments | tojson }}
77
+ {%- endif %}
78
+ {{- '}\n</tool_call>' }}
79
+ {%- endfor %}
80
+ {%- endif %}
81
+ {{- '<|im_end|>\n' }}
82
+ {% endgeneration %}
83
+ {%- elif message.role == "tool" %}
84
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
85
+ {{- '<|im_start|>user' }}
86
+ {%- endif %}
87
+ {{- '\n<tool_response>\n' }}
88
+ {{- content }}
89
+ {{- '\n</tool_response>' }}
90
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
91
+ {{- '<|im_end|>\n' }}
92
+ {%- endif %}
93
+ {%- endif %}
94
+ {%- endfor %}
95
+ {%- if add_generation_prompt %}
96
+ {{- '<|im_start|>assistant\n' }}
97
+ {%- if enable_thinking is defined and enable_thinking is false %}
98
+ {{- '<think>\n\n</think>\n\n' }}
99
+ {%- endif %}
100
+ {%- endif %}
config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151645,
9
+ "head_dim": 128,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 2048,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 6144,
14
+ "layer_types": [
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
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention"
43
+ ],
44
+ "max_position_embeddings": 40960,
45
+ "max_window_layers": 28,
46
+ "model_type": "qwen3",
47
+ "num_attention_heads": 16,
48
+ "num_hidden_layers": 28,
49
+ "num_key_value_heads": 8,
50
+ "rms_norm_eps": 1e-06,
51
+ "rope_scaling": null,
52
+ "rope_theta": 1000000,
53
+ "sliding_window": null,
54
+ "tie_word_embeddings": true,
55
+ "torch_dtype": "bfloat16",
56
+ "transformers_version": "4.55.0",
57
+ "use_cache": false,
58
+ "use_sliding_window": false,
59
+ "vocab_size": 151936
60
+ }
generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "pad_token_id": 151643,
9
+ "temperature": 0.6,
10
+ "top_k": 20,
11
+ "top_p": 0.95,
12
+ "transformers_version": "4.55.0"
13
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c51ce2cb6bf69f96633cce65d321f6e767bf121cecd033069e90a149c3f7a76
3
+ size 3441185608
special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
3
+ size 11422654
tokenizer_config.json ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ },
181
+ "151665": {
182
+ "content": "<tool_response>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": false
188
+ },
189
+ "151666": {
190
+ "content": "</tool_response>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": false
196
+ },
197
+ "151667": {
198
+ "content": "<think>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": false
204
+ },
205
+ "151668": {
206
+ "content": "</think>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": false
212
+ }
213
+ },
214
+ "additional_special_tokens": [
215
+ "<|im_start|>",
216
+ "<|im_end|>",
217
+ "<|object_ref_start|>",
218
+ "<|object_ref_end|>",
219
+ "<|box_start|>",
220
+ "<|box_end|>",
221
+ "<|quad_start|>",
222
+ "<|quad_end|>",
223
+ "<|vision_start|>",
224
+ "<|vision_end|>",
225
+ "<|vision_pad|>",
226
+ "<|image_pad|>",
227
+ "<|video_pad|>"
228
+ ],
229
+ "bos_token": null,
230
+ "clean_up_tokenization_spaces": false,
231
+ "eos_token": "<|im_end|>",
232
+ "errors": "replace",
233
+ "extra_special_tokens": {},
234
+ "model_max_length": 131072,
235
+ "pad_token": "<|endoftext|>",
236
+ "split_special_tokens": false,
237
+ "tokenizer_class": "Qwen2Tokenizer",
238
+ "unk_token": null
239
+ }
trainer_state.json ADDED
@@ -0,0 +1,1414 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.0019372336303758234,
6
+ "eval_steps": 500,
7
+ "global_step": 80,
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": 0.0,
19
+ "completions/max_length": 5021.0,
20
+ "completions/max_terminated_length": 5021.0,
21
+ "completions/mean_length": 3011.5625,
22
+ "completions/mean_terminated_length": 3011.5625,
23
+ "completions/min_length": 1324.0,
24
+ "completions/min_terminated_length": 1324.0,
25
+ "epoch": 2.4215420379697792e-05,
26
+ "grad_norm": 0.0013862343573498733,
27
+ "kl": 0.000823974609375,
28
+ "learning_rate": 0.0,
29
+ "loss": -0.0001,
30
+ "num_tokens": 227146.0,
31
+ "reward": 0.8549610376358032,
32
+ "reward_std": 0.021876953542232513,
33
+ "rewards/avg_thinking_length_func": 157.7991485595703,
34
+ "rewards/correct_answer_reward_func": 0.75,
35
+ "rewards/efficient_thinking_reward_func": 0.9122447304419112,
36
+ "rewards/format_reward_func": 1.0,
37
+ "rewards/num_xml_reward_func": 1.7957969903945923,
38
+ "rewards/tool_execution_reward_func": 1.9947917461395264,
39
+ "rewards/visit_tool_reward_func": 0.9724012017250061,
40
+ "step": 1
41
+ },
42
+ {
43
+ "clip_ratio/high_max": 0.0,
44
+ "clip_ratio/high_mean": 0.0,
45
+ "clip_ratio/low_mean": 0.0,
46
+ "clip_ratio/low_min": 0.0,
47
+ "clip_ratio/region_mean": 0.0,
48
+ "epoch": 4.8430840759395585e-05,
49
+ "grad_norm": 0.0013858304532204059,
50
+ "kl": 0.000823974609375,
51
+ "learning_rate": 6.25e-08,
52
+ "loss": -0.0001,
53
+ "step": 2
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
+ "epoch": 7.264626113909338e-05,
62
+ "grad_norm": 0.0013865629989697915,
63
+ "kl": 0.000820159912109375,
64
+ "learning_rate": 1.25e-07,
65
+ "loss": -0.0001,
66
+ "step": 3
67
+ },
68
+ {
69
+ "clip_ratio/high_max": 0.0,
70
+ "clip_ratio/high_mean": 0.0,
71
+ "clip_ratio/low_mean": 0.0,
72
+ "clip_ratio/low_min": 0.0,
73
+ "clip_ratio/region_mean": 0.0,
74
+ "epoch": 9.686168151879117e-05,
75
+ "grad_norm": 0.0013802536996446572,
76
+ "kl": 0.0008420944213867188,
77
+ "learning_rate": 1.875e-07,
78
+ "loss": -0.0001,
79
+ "step": 4
80
+ },
81
+ {
82
+ "clip_ratio/high_max": 0.0,
83
+ "clip_ratio/high_mean": 0.0,
84
+ "clip_ratio/low_mean": 0.0,
85
+ "clip_ratio/low_min": 0.0,
86
+ "clip_ratio/region_mean": 0.0,
87
+ "completions/clipped_ratio": 0.0,
88
+ "completions/max_length": 8221.0,
89
+ "completions/max_terminated_length": 8221.0,
90
+ "completions/mean_length": 5079.5625,
91
+ "completions/mean_terminated_length": 5079.5625,
92
+ "completions/min_length": 2306.0,
93
+ "completions/min_terminated_length": 2306.0,
94
+ "epoch": 0.00012107710189848896,
95
+ "grad_norm": 0.03869369457145168,
96
+ "kl": 0.0009288787841796875,
97
+ "learning_rate": 2.5e-07,
98
+ "loss": 0.0052,
99
+ "num_tokens": 580872.0,
100
+ "reward": 0.30476048588752747,
101
+ "reward_std": 0.4756142497062683,
102
+ "rewards/avg_thinking_length_func": 195.94981384277344,
103
+ "rewards/correct_answer_reward_func": 0.3125,
104
+ "rewards/efficient_thinking_reward_func": 0.8946171495693498,
105
+ "rewards/format_reward_func": 1.0,
106
+ "rewards/num_xml_reward_func": 1.3311374187469482,
107
+ "rewards/tool_execution_reward_func": 2.0,
108
+ "rewards/visit_tool_reward_func": 1.0090813636779785,
109
+ "step": 5
110
+ },
111
+ {
112
+ "clip_ratio/high_max": 0.0,
113
+ "clip_ratio/high_mean": 0.0,
114
+ "clip_ratio/low_mean": 0.0,
115
+ "clip_ratio/low_min": 0.0,
116
+ "clip_ratio/region_mean": 0.0,
117
+ "epoch": 0.00014529252227818675,
118
+ "grad_norm": 0.03816851177241048,
119
+ "kl": 0.0009202957153320312,
120
+ "learning_rate": 3.1249999999999997e-07,
121
+ "loss": 0.0052,
122
+ "step": 6
123
+ },
124
+ {
125
+ "clip_ratio/high_max": 0.0,
126
+ "clip_ratio/high_mean": 0.0,
127
+ "clip_ratio/low_mean": 0.0,
128
+ "clip_ratio/low_min": 0.0,
129
+ "clip_ratio/region_mean": 0.0,
130
+ "epoch": 0.00016950794265788453,
131
+ "grad_norm": 0.03841593140613752,
132
+ "kl": 0.0009307861328125,
133
+ "learning_rate": 3.75e-07,
134
+ "loss": 0.0052,
135
+ "step": 7
136
+ },
137
+ {
138
+ "clip_ratio/high_max": 0.0,
139
+ "clip_ratio/high_mean": 0.0,
140
+ "clip_ratio/low_mean": 0.0,
141
+ "clip_ratio/low_min": 0.0,
142
+ "clip_ratio/region_mean": 0.0,
143
+ "epoch": 0.00019372336303758234,
144
+ "grad_norm": 0.03873458948340372,
145
+ "kl": 0.000919342041015625,
146
+ "learning_rate": 4.375e-07,
147
+ "loss": 0.0052,
148
+ "step": 8
149
+ },
150
+ {
151
+ "clip_ratio/high_max": 0.0,
152
+ "clip_ratio/high_mean": 0.0,
153
+ "clip_ratio/low_mean": 0.0,
154
+ "clip_ratio/low_min": 0.0,
155
+ "clip_ratio/region_mean": 0.0,
156
+ "completions/clipped_ratio": 0.0,
157
+ "completions/max_length": 8205.0,
158
+ "completions/max_terminated_length": 8205.0,
159
+ "completions/mean_length": 4930.875,
160
+ "completions/mean_terminated_length": 4930.875,
161
+ "completions/min_length": 2085.0,
162
+ "completions/min_terminated_length": 2085.0,
163
+ "epoch": 0.00021793878341728012,
164
+ "grad_norm": 0.027339756517219562,
165
+ "kl": 0.0009183883666992188,
166
+ "learning_rate": 5e-07,
167
+ "loss": 0.0039,
168
+ "num_tokens": 926931.0,
169
+ "reward": 0.41861122846603394,
170
+ "reward_std": 0.399558424949646,
171
+ "rewards/avg_thinking_length_func": 189.7052764892578,
172
+ "rewards/correct_answer_reward_func": 0.40625,
173
+ "rewards/efficient_thinking_reward_func": 0.9012974528882096,
174
+ "rewards/format_reward_func": 1.0,
175
+ "rewards/num_xml_reward_func": 1.572190284729004,
176
+ "rewards/tool_execution_reward_func": 2.0,
177
+ "rewards/visit_tool_reward_func": 1.0377196073532104,
178
+ "step": 9
179
+ },
180
+ {
181
+ "clip_ratio/high_max": 0.0,
182
+ "clip_ratio/high_mean": 0.0,
183
+ "clip_ratio/low_mean": 0.0,
184
+ "clip_ratio/low_min": 0.0,
185
+ "clip_ratio/region_mean": 0.0,
186
+ "epoch": 0.00024215420379697792,
187
+ "grad_norm": 0.027389009217824802,
188
+ "kl": 0.0009317398071289062,
189
+ "learning_rate": 5.625e-07,
190
+ "loss": 0.0039,
191
+ "step": 10
192
+ },
193
+ {
194
+ "clip_ratio/high_max": 0.0,
195
+ "clip_ratio/high_mean": 0.0,
196
+ "clip_ratio/low_mean": 0.0,
197
+ "clip_ratio/low_min": 0.0,
198
+ "clip_ratio/region_mean": 0.0,
199
+ "epoch": 0.0002663696241766757,
200
+ "grad_norm": 0.027404975057647246,
201
+ "kl": 0.0009403228759765625,
202
+ "learning_rate": 6.249999999999999e-07,
203
+ "loss": 0.0039,
204
+ "step": 11
205
+ },
206
+ {
207
+ "clip_ratio/high_max": 0.0,
208
+ "clip_ratio/high_mean": 0.0,
209
+ "clip_ratio/low_mean": 0.0,
210
+ "clip_ratio/low_min": 0.0,
211
+ "clip_ratio/region_mean": 0.0,
212
+ "epoch": 0.0002905850445563735,
213
+ "grad_norm": 0.027663414412920105,
214
+ "kl": 0.0010051727294921875,
215
+ "learning_rate": 6.875e-07,
216
+ "loss": 0.0039,
217
+ "step": 12
218
+ },
219
+ {
220
+ "clip_ratio/high_max": 0.0,
221
+ "clip_ratio/high_mean": 0.0,
222
+ "clip_ratio/low_mean": 0.0,
223
+ "clip_ratio/low_min": 0.0,
224
+ "clip_ratio/region_mean": 0.0,
225
+ "completions/clipped_ratio": 0.0,
226
+ "completions/max_length": 8889.0,
227
+ "completions/max_terminated_length": 8889.0,
228
+ "completions/mean_length": 5363.5,
229
+ "completions/mean_terminated_length": 5363.5,
230
+ "completions/min_length": 1743.0,
231
+ "completions/min_terminated_length": 1743.0,
232
+ "epoch": 0.0003148004649360713,
233
+ "grad_norm": 0.0407893748945086,
234
+ "kl": 0.0009937286376953125,
235
+ "learning_rate": 7.5e-07,
236
+ "loss": -0.0002,
237
+ "num_tokens": 1294834.0,
238
+ "reward": 0.3248969614505768,
239
+ "reward_std": 0.6025969982147217,
240
+ "rewards/avg_thinking_length_func": 197.81080627441406,
241
+ "rewards/correct_answer_reward_func": 0.375,
242
+ "rewards/efficient_thinking_reward_func": 0.8764530262818557,
243
+ "rewards/format_reward_func": 1.0,
244
+ "rewards/num_xml_reward_func": 1.5131020545959473,
245
+ "rewards/tool_execution_reward_func": 1.9973957538604736,
246
+ "rewards/visit_tool_reward_func": 1.029296636581421,
247
+ "step": 13
248
+ },
249
+ {
250
+ "clip_ratio/high_max": 0.0,
251
+ "clip_ratio/high_mean": 0.0,
252
+ "clip_ratio/low_mean": 0.0,
253
+ "clip_ratio/low_min": 0.0,
254
+ "clip_ratio/region_mean": 0.0,
255
+ "epoch": 0.00033901588531576906,
256
+ "grad_norm": 0.04111370827114344,
257
+ "kl": 0.0009851455688476562,
258
+ "learning_rate": 8.125e-07,
259
+ "loss": -0.0002,
260
+ "step": 14
261
+ },
262
+ {
263
+ "clip_ratio/high_max": 0.0,
264
+ "clip_ratio/high_mean": 0.0,
265
+ "clip_ratio/low_mean": 0.0,
266
+ "clip_ratio/low_min": 0.0,
267
+ "clip_ratio/region_mean": 0.0,
268
+ "epoch": 0.00036323130569546687,
269
+ "grad_norm": 0.040997833074323406,
270
+ "kl": 0.0010662078857421875,
271
+ "learning_rate": 8.75e-07,
272
+ "loss": -0.0002,
273
+ "step": 15
274
+ },
275
+ {
276
+ "clip_ratio/high_max": 0.0,
277
+ "clip_ratio/high_mean": 0.0,
278
+ "clip_ratio/low_mean": 0.0,
279
+ "clip_ratio/low_min": 0.0,
280
+ "clip_ratio/region_mean": 0.0,
281
+ "epoch": 0.0003874467260751647,
282
+ "grad_norm": 0.04130516332775726,
283
+ "kl": 0.0011501312255859375,
284
+ "learning_rate": 9.374999999999999e-07,
285
+ "loss": -0.0002,
286
+ "step": 16
287
+ },
288
+ {
289
+ "clip_ratio/high_max": 0.0,
290
+ "clip_ratio/high_mean": 0.0,
291
+ "clip_ratio/low_mean": 0.0,
292
+ "clip_ratio/low_min": 0.0,
293
+ "clip_ratio/region_mean": 0.0,
294
+ "completions/clipped_ratio": 0.0,
295
+ "completions/max_length": 8681.0,
296
+ "completions/max_terminated_length": 8681.0,
297
+ "completions/mean_length": 5228.8125,
298
+ "completions/mean_terminated_length": 5228.8125,
299
+ "completions/min_length": 2233.0,
300
+ "completions/min_terminated_length": 2233.0,
301
+ "epoch": 0.0004116621464548625,
302
+ "grad_norm": 0.03978610475757746,
303
+ "kl": 0.001308441162109375,
304
+ "learning_rate": 1e-06,
305
+ "loss": 0.006,
306
+ "num_tokens": 1648672.0,
307
+ "reward": 0.34718748927116394,
308
+ "reward_std": 0.5881420969963074,
309
+ "rewards/avg_thinking_length_func": 195.60931396484375,
310
+ "rewards/correct_answer_reward_func": 0.34375,
311
+ "rewards/efficient_thinking_reward_func": 0.8885279571058924,
312
+ "rewards/format_reward_func": 1.0,
313
+ "rewards/num_xml_reward_func": 1.63374924659729,
314
+ "rewards/tool_execution_reward_func": 2.0,
315
+ "rewards/visit_tool_reward_func": 1.035430908203125,
316
+ "step": 17
317
+ },
318
+ {
319
+ "clip_ratio/high_max": 0.0,
320
+ "clip_ratio/high_mean": 0.0,
321
+ "clip_ratio/low_mean": 0.0,
322
+ "clip_ratio/low_min": 0.0,
323
+ "clip_ratio/region_mean": 0.0,
324
+ "epoch": 0.00043587756683456023,
325
+ "grad_norm": 0.04006457357588843,
326
+ "kl": 0.001514434814453125,
327
+ "learning_rate": 1.0625e-06,
328
+ "loss": 0.006,
329
+ "step": 18
330
+ },
331
+ {
332
+ "clip_ratio/high_max": 0.0,
333
+ "clip_ratio/high_mean": 0.0,
334
+ "clip_ratio/low_mean": 0.0,
335
+ "clip_ratio/low_min": 0.0,
336
+ "clip_ratio/region_mean": 0.0,
337
+ "epoch": 0.00046009298721425804,
338
+ "grad_norm": 0.04053645195569142,
339
+ "kl": 0.0020732879638671875,
340
+ "learning_rate": 1.125e-06,
341
+ "loss": 0.006,
342
+ "step": 19
343
+ },
344
+ {
345
+ "clip_ratio/high_max": 0.0,
346
+ "clip_ratio/high_mean": 0.0,
347
+ "clip_ratio/low_mean": 0.0,
348
+ "clip_ratio/low_min": 0.0,
349
+ "clip_ratio/region_mean": 0.0,
350
+ "epoch": 0.00048430840759395585,
351
+ "grad_norm": 0.040941267103466746,
352
+ "kl": 0.002826690673828125,
353
+ "learning_rate": 1.1874999999999999e-06,
354
+ "loss": 0.006,
355
+ "step": 20
356
+ },
357
+ {
358
+ "clip_ratio/high_max": 0.0,
359
+ "clip_ratio/high_mean": 0.0,
360
+ "clip_ratio/low_mean": 0.0,
361
+ "clip_ratio/low_min": 0.0,
362
+ "clip_ratio/region_mean": 0.0,
363
+ "completions/clipped_ratio": 0.0,
364
+ "completions/max_length": 8595.0,
365
+ "completions/max_terminated_length": 8595.0,
366
+ "completions/mean_length": 5146.78125,
367
+ "completions/mean_terminated_length": 5146.78125,
368
+ "completions/min_length": 2093.0,
369
+ "completions/min_terminated_length": 2093.0,
370
+ "epoch": 0.0005085238279736536,
371
+ "grad_norm": 0.034201382636368564,
372
+ "kl": 0.00348663330078125,
373
+ "learning_rate": 1.2499999999999999e-06,
374
+ "loss": 0.0012,
375
+ "num_tokens": 2015057.0,
376
+ "reward": 0.3354427218437195,
377
+ "reward_std": 0.5395293831825256,
378
+ "rewards/avg_thinking_length_func": 181.95034790039062,
379
+ "rewards/correct_answer_reward_func": 0.4375,
380
+ "rewards/efficient_thinking_reward_func": 0.8707138202597767,
381
+ "rewards/format_reward_func": 0.9997023940086365,
382
+ "rewards/num_xml_reward_func": 1.4494500160217285,
383
+ "rewards/tool_execution_reward_func": 2.0,
384
+ "rewards/visit_tool_reward_func": 1.0431455373764038,
385
+ "step": 21
386
+ },
387
+ {
388
+ "clip_ratio/high_max": 0.0,
389
+ "clip_ratio/high_mean": 0.0,
390
+ "clip_ratio/low_mean": 0.0,
391
+ "clip_ratio/low_min": 0.0,
392
+ "clip_ratio/region_mean": 0.0,
393
+ "epoch": 0.0005327392483533514,
394
+ "grad_norm": 0.03384948845768777,
395
+ "kl": 0.00420379638671875,
396
+ "learning_rate": 1.3125e-06,
397
+ "loss": 0.0012,
398
+ "step": 22
399
+ },
400
+ {
401
+ "clip_ratio/high_max": 0.0,
402
+ "clip_ratio/high_mean": 0.0,
403
+ "clip_ratio/low_mean": 0.0,
404
+ "clip_ratio/low_min": 0.0,
405
+ "clip_ratio/region_mean": 0.0,
406
+ "epoch": 0.0005569546687330492,
407
+ "grad_norm": 0.03413362282042912,
408
+ "kl": 0.0050048828125,
409
+ "learning_rate": 1.375e-06,
410
+ "loss": 0.0012,
411
+ "step": 23
412
+ },
413
+ {
414
+ "clip_ratio/high_max": 0.0,
415
+ "clip_ratio/high_mean": 0.0,
416
+ "clip_ratio/low_mean": 0.0,
417
+ "clip_ratio/low_min": 0.0,
418
+ "clip_ratio/region_mean": 0.0,
419
+ "epoch": 0.000581170089112747,
420
+ "grad_norm": 0.03410120905325483,
421
+ "kl": 0.0059661865234375,
422
+ "learning_rate": 1.4375e-06,
423
+ "loss": 0.0012,
424
+ "step": 24
425
+ },
426
+ {
427
+ "clip_ratio/high_max": 0.0,
428
+ "clip_ratio/high_mean": 0.0,
429
+ "clip_ratio/low_mean": 0.0,
430
+ "clip_ratio/low_min": 0.0,
431
+ "clip_ratio/region_mean": 0.0,
432
+ "completions/clipped_ratio": 0.0,
433
+ "completions/max_length": 8285.0,
434
+ "completions/max_terminated_length": 8285.0,
435
+ "completions/mean_length": 5087.59375,
436
+ "completions/mean_terminated_length": 5087.59375,
437
+ "completions/min_length": 2272.0,
438
+ "completions/min_terminated_length": 2272.0,
439
+ "epoch": 0.0006053855094924448,
440
+ "grad_norm": 0.044445551609988844,
441
+ "kl": 0.00791168212890625,
442
+ "learning_rate": 1.5e-06,
443
+ "loss": 0.0078,
444
+ "num_tokens": 2383540.0,
445
+ "reward": 0.3267378509044647,
446
+ "reward_std": 0.4416780471801758,
447
+ "rewards/avg_thinking_length_func": 174.10765075683594,
448
+ "rewards/correct_answer_reward_func": 0.28125,
449
+ "rewards/efficient_thinking_reward_func": 0.9263086220559518,
450
+ "rewards/format_reward_func": 1.0,
451
+ "rewards/num_xml_reward_func": 1.755671501159668,
452
+ "rewards/tool_execution_reward_func": 1.9970238208770752,
453
+ "rewards/visit_tool_reward_func": 1.0175025463104248,
454
+ "step": 25
455
+ },
456
+ {
457
+ "clip_ratio/high_max": 0.0,
458
+ "clip_ratio/high_mean": 0.0,
459
+ "clip_ratio/low_mean": 0.0,
460
+ "clip_ratio/low_min": 0.0,
461
+ "clip_ratio/region_mean": 0.0,
462
+ "epoch": 0.0006296009298721426,
463
+ "grad_norm": 0.04825803348555062,
464
+ "kl": 0.012359619140625,
465
+ "learning_rate": 1.5624999999999999e-06,
466
+ "loss": 0.0078,
467
+ "step": 26
468
+ },
469
+ {
470
+ "clip_ratio/high_max": 0.0,
471
+ "clip_ratio/high_mean": 0.0,
472
+ "clip_ratio/low_mean": 0.0,
473
+ "clip_ratio/low_min": 0.0,
474
+ "clip_ratio/region_mean": 0.0,
475
+ "epoch": 0.0006538163502518403,
476
+ "grad_norm": 0.05116686338837774,
477
+ "kl": 0.021484375,
478
+ "learning_rate": 1.625e-06,
479
+ "loss": 0.0078,
480
+ "step": 27
481
+ },
482
+ {
483
+ "clip_ratio/high_max": 0.0,
484
+ "clip_ratio/high_mean": 0.0,
485
+ "clip_ratio/low_mean": 0.0,
486
+ "clip_ratio/low_min": 0.0,
487
+ "clip_ratio/region_mean": 0.0,
488
+ "epoch": 0.0006780317706315381,
489
+ "grad_norm": 0.05316346328608609,
490
+ "kl": 0.03985595703125,
491
+ "learning_rate": 1.6875e-06,
492
+ "loss": 0.0078,
493
+ "step": 28
494
+ },
495
+ {
496
+ "clip_ratio/high_max": 0.0,
497
+ "clip_ratio/high_mean": 0.0,
498
+ "clip_ratio/low_mean": 0.0,
499
+ "clip_ratio/low_min": 0.0,
500
+ "clip_ratio/region_mean": 0.0,
501
+ "completions/clipped_ratio": 0.0,
502
+ "completions/max_length": 7189.0,
503
+ "completions/max_terminated_length": 7189.0,
504
+ "completions/mean_length": 4996.46875,
505
+ "completions/mean_terminated_length": 4996.46875,
506
+ "completions/min_length": 2314.0,
507
+ "completions/min_terminated_length": 2314.0,
508
+ "epoch": 0.0007022471910112359,
509
+ "grad_norm": 0.04891502789141398,
510
+ "kl": 0.04736328125,
511
+ "learning_rate": 1.75e-06,
512
+ "loss": 0.0067,
513
+ "num_tokens": 2761999.0,
514
+ "reward": 0.41611552238464355,
515
+ "reward_std": 0.456102192401886,
516
+ "rewards/avg_thinking_length_func": 165.63626098632812,
517
+ "rewards/correct_answer_reward_func": 0.40625,
518
+ "rewards/efficient_thinking_reward_func": 0.9453047237018808,
519
+ "rewards/format_reward_func": 1.0,
520
+ "rewards/num_xml_reward_func": 1.5178555250167847,
521
+ "rewards/tool_execution_reward_func": 1.9925000667572021,
522
+ "rewards/visit_tool_reward_func": 1.0254861116409302,
523
+ "step": 29
524
+ },
525
+ {
526
+ "clip_ratio/high_max": 0.0,
527
+ "clip_ratio/high_mean": 0.0,
528
+ "clip_ratio/low_mean": 0.0,
529
+ "clip_ratio/low_min": 0.0,
530
+ "clip_ratio/region_mean": 0.0,
531
+ "epoch": 0.0007264626113909337,
532
+ "grad_norm": 0.05075936350489512,
533
+ "kl": 0.07891845703125,
534
+ "learning_rate": 1.8125e-06,
535
+ "loss": 0.0067,
536
+ "step": 30
537
+ },
538
+ {
539
+ "clip_ratio/high_max": 0.0,
540
+ "clip_ratio/high_mean": 0.0,
541
+ "clip_ratio/low_mean": 0.0,
542
+ "clip_ratio/low_min": 0.0,
543
+ "clip_ratio/region_mean": 0.0,
544
+ "epoch": 0.0007506780317706315,
545
+ "grad_norm": 0.05378691343356215,
546
+ "kl": 0.1292724609375,
547
+ "learning_rate": 1.8749999999999998e-06,
548
+ "loss": 0.0068,
549
+ "step": 31
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
+ "epoch": 0.0007748934521503294,
558
+ "grad_norm": 0.06810533074784274,
559
+ "kl": 0.19677734375,
560
+ "learning_rate": 1.9375e-06,
561
+ "loss": 0.0069,
562
+ "step": 32
563
+ },
564
+ {
565
+ "clip_ratio/high_max": 0.0,
566
+ "clip_ratio/high_mean": 0.0,
567
+ "clip_ratio/low_mean": 0.0,
568
+ "clip_ratio/low_min": 0.0,
569
+ "clip_ratio/region_mean": 0.0,
570
+ "completions/clipped_ratio": 0.0,
571
+ "completions/max_length": 6880.0,
572
+ "completions/max_terminated_length": 6880.0,
573
+ "completions/mean_length": 5091.3125,
574
+ "completions/mean_terminated_length": 5091.3125,
575
+ "completions/min_length": 2317.0,
576
+ "completions/min_terminated_length": 2317.0,
577
+ "epoch": 0.0007991088725300272,
578
+ "grad_norm": 0.1762564569620267,
579
+ "kl": 0.204345703125,
580
+ "learning_rate": 2e-06,
581
+ "loss": 0.0021,
582
+ "num_tokens": 3193245.0,
583
+ "reward": 0.606549859046936,
584
+ "reward_std": 0.42073318362236023,
585
+ "rewards/avg_thinking_length_func": 136.30906677246094,
586
+ "rewards/correct_answer_reward_func": 0.5625,
587
+ "rewards/efficient_thinking_reward_func": 0.9376242936192475,
588
+ "rewards/format_reward_func": 1.0,
589
+ "rewards/num_xml_reward_func": 1.400514841079712,
590
+ "rewards/tool_execution_reward_func": 2.0,
591
+ "rewards/visit_tool_reward_func": 1.0149691104888916,
592
+ "step": 33
593
+ },
594
+ {
595
+ "clip_ratio/high_max": 0.0,
596
+ "clip_ratio/high_mean": 0.0,
597
+ "clip_ratio/low_mean": 0.0,
598
+ "clip_ratio/low_min": 0.0,
599
+ "clip_ratio/region_mean": 0.0,
600
+ "epoch": 0.000823324292909725,
601
+ "grad_norm": 0.09601002970053912,
602
+ "kl": 0.177001953125,
603
+ "learning_rate": 2e-06,
604
+ "loss": 0.0021,
605
+ "step": 34
606
+ },
607
+ {
608
+ "clip_ratio/high_max": 0.0,
609
+ "clip_ratio/high_mean": 0.0,
610
+ "clip_ratio/low_mean": 0.0,
611
+ "clip_ratio/low_min": 0.0,
612
+ "clip_ratio/region_mean": 0.0,
613
+ "epoch": 0.0008475397132894227,
614
+ "grad_norm": 0.059153354466426514,
615
+ "kl": 0.1494140625,
616
+ "learning_rate": 2e-06,
617
+ "loss": 0.002,
618
+ "step": 35
619
+ },
620
+ {
621
+ "clip_ratio/high_max": 0.0,
622
+ "clip_ratio/high_mean": 0.0,
623
+ "clip_ratio/low_mean": 0.0,
624
+ "clip_ratio/low_min": 0.0,
625
+ "clip_ratio/region_mean": 0.0,
626
+ "epoch": 0.0008717551336691205,
627
+ "grad_norm": 0.04692926604339145,
628
+ "kl": 0.134033203125,
629
+ "learning_rate": 2e-06,
630
+ "loss": 0.002,
631
+ "step": 36
632
+ },
633
+ {
634
+ "clip_ratio/high_max": 0.0,
635
+ "clip_ratio/high_mean": 0.0,
636
+ "clip_ratio/low_mean": 0.0,
637
+ "clip_ratio/low_min": 0.0,
638
+ "clip_ratio/region_mean": 0.0,
639
+ "completions/clipped_ratio": 0.0,
640
+ "completions/max_length": 6702.0,
641
+ "completions/max_terminated_length": 6702.0,
642
+ "completions/mean_length": 4464.4375,
643
+ "completions/mean_terminated_length": 4464.4375,
644
+ "completions/min_length": 2517.0,
645
+ "completions/min_terminated_length": 2517.0,
646
+ "epoch": 0.0008959705540488183,
647
+ "grad_norm": 0.04156023447496563,
648
+ "kl": 0.089111328125,
649
+ "learning_rate": 2e-06,
650
+ "loss": 0.0009,
651
+ "num_tokens": 3614725.0,
652
+ "reward": 0.5230777859687805,
653
+ "reward_std": 0.4564175009727478,
654
+ "rewards/avg_thinking_length_func": 113.61294555664062,
655
+ "rewards/correct_answer_reward_func": 0.5,
656
+ "rewards/efficient_thinking_reward_func": 0.8956072636293169,
657
+ "rewards/format_reward_func": 1.0,
658
+ "rewards/num_xml_reward_func": 1.4630262851715088,
659
+ "rewards/tool_execution_reward_func": 1.9854960441589355,
660
+ "rewards/visit_tool_reward_func": 0.9511741399765015,
661
+ "step": 37
662
+ },
663
+ {
664
+ "clip_ratio/high_max": 0.0,
665
+ "clip_ratio/high_mean": 0.0,
666
+ "clip_ratio/low_mean": 0.0,
667
+ "clip_ratio/low_min": 0.0,
668
+ "clip_ratio/region_mean": 0.0,
669
+ "epoch": 0.0009201859744285161,
670
+ "grad_norm": 0.030847266935293525,
671
+ "kl": 0.0772705078125,
672
+ "learning_rate": 2e-06,
673
+ "loss": 0.0009,
674
+ "step": 38
675
+ },
676
+ {
677
+ "clip_ratio/high_max": 0.0,
678
+ "clip_ratio/high_mean": 0.0,
679
+ "clip_ratio/low_mean": 0.0,
680
+ "clip_ratio/low_min": 0.0,
681
+ "clip_ratio/region_mean": 0.0,
682
+ "epoch": 0.0009444013948082139,
683
+ "grad_norm": 0.029857330293238966,
684
+ "kl": 0.0762939453125,
685
+ "learning_rate": 2e-06,
686
+ "loss": 0.0009,
687
+ "step": 39
688
+ },
689
+ {
690
+ "clip_ratio/high_max": 0.0,
691
+ "clip_ratio/high_mean": 0.0,
692
+ "clip_ratio/low_mean": 0.0,
693
+ "clip_ratio/low_min": 0.0,
694
+ "clip_ratio/region_mean": 0.0,
695
+ "epoch": 0.0009686168151879117,
696
+ "grad_norm": 0.029616934835158485,
697
+ "kl": 0.0780029296875,
698
+ "learning_rate": 2e-06,
699
+ "loss": 0.0009,
700
+ "step": 40
701
+ },
702
+ {
703
+ "clip_ratio/high_max": 0.0,
704
+ "clip_ratio/high_mean": 0.0,
705
+ "clip_ratio/low_mean": 0.0,
706
+ "clip_ratio/low_min": 0.0,
707
+ "clip_ratio/region_mean": 0.0,
708
+ "completions/clipped_ratio": 0.0,
709
+ "completions/max_length": 5952.0,
710
+ "completions/max_terminated_length": 5952.0,
711
+ "completions/mean_length": 4219.25,
712
+ "completions/mean_terminated_length": 4219.25,
713
+ "completions/min_length": 2282.0,
714
+ "completions/min_terminated_length": 2282.0,
715
+ "epoch": 0.0009928322355676094,
716
+ "grad_norm": 0.027887879707077786,
717
+ "kl": 0.0771484375,
718
+ "learning_rate": 2e-06,
719
+ "loss": 0.0014,
720
+ "num_tokens": 4009367.0,
721
+ "reward": 0.6270047426223755,
722
+ "reward_std": 0.432157427072525,
723
+ "rewards/avg_thinking_length_func": 115.17754364013672,
724
+ "rewards/correct_answer_reward_func": 0.59375,
725
+ "rewards/efficient_thinking_reward_func": 0.9136282100129471,
726
+ "rewards/format_reward_func": 1.0,
727
+ "rewards/num_xml_reward_func": 1.6435627937316895,
728
+ "rewards/tool_execution_reward_func": 1.9891107082366943,
729
+ "rewards/visit_tool_reward_func": 0.9315186738967896,
730
+ "step": 41
731
+ },
732
+ {
733
+ "clip_ratio/high_max": 0.0,
734
+ "clip_ratio/high_mean": 0.0,
735
+ "clip_ratio/low_mean": 0.0,
736
+ "clip_ratio/low_min": 0.0,
737
+ "clip_ratio/region_mean": 0.0,
738
+ "epoch": 0.0010170476559473072,
739
+ "grad_norm": 0.02835635253506866,
740
+ "kl": 0.0794677734375,
741
+ "learning_rate": 2e-06,
742
+ "loss": 0.0014,
743
+ "step": 42
744
+ },
745
+ {
746
+ "clip_ratio/high_max": 0.0,
747
+ "clip_ratio/high_mean": 0.0,
748
+ "clip_ratio/low_mean": 0.0,
749
+ "clip_ratio/low_min": 0.0,
750
+ "clip_ratio/region_mean": 0.0,
751
+ "epoch": 0.001041263076327005,
752
+ "grad_norm": 0.028940878534019835,
753
+ "kl": 0.083740234375,
754
+ "learning_rate": 2e-06,
755
+ "loss": 0.0014,
756
+ "step": 43
757
+ },
758
+ {
759
+ "clip_ratio/high_max": 0.0,
760
+ "clip_ratio/high_mean": 0.0,
761
+ "clip_ratio/low_mean": 0.0,
762
+ "clip_ratio/low_min": 0.0,
763
+ "clip_ratio/region_mean": 0.0,
764
+ "epoch": 0.0010654784967067028,
765
+ "grad_norm": 0.029849752671864262,
766
+ "kl": 0.08837890625,
767
+ "learning_rate": 2e-06,
768
+ "loss": 0.0014,
769
+ "step": 44
770
+ },
771
+ {
772
+ "clip_ratio/high_max": 0.0,
773
+ "clip_ratio/high_mean": 0.0,
774
+ "clip_ratio/low_mean": 0.0,
775
+ "clip_ratio/low_min": 0.0,
776
+ "clip_ratio/region_mean": 0.0,
777
+ "completions/clipped_ratio": 0.0,
778
+ "completions/max_length": 6019.0,
779
+ "completions/max_terminated_length": 6019.0,
780
+ "completions/mean_length": 4090.75,
781
+ "completions/mean_terminated_length": 4090.75,
782
+ "completions/min_length": 2240.0,
783
+ "completions/min_terminated_length": 2240.0,
784
+ "epoch": 0.0010896939170864006,
785
+ "grad_norm": 0.02660440382201049,
786
+ "kl": 0.09228515625,
787
+ "learning_rate": 2e-06,
788
+ "loss": -0.0002,
789
+ "num_tokens": 4366159.0,
790
+ "reward": 0.6755548715591431,
791
+ "reward_std": 0.47086238861083984,
792
+ "rewards/avg_thinking_length_func": 125.40975189208984,
793
+ "rewards/correct_answer_reward_func": 0.6875,
794
+ "rewards/efficient_thinking_reward_func": 0.9314815674427694,
795
+ "rewards/format_reward_func": 1.0,
796
+ "rewards/num_xml_reward_func": 1.7004401683807373,
797
+ "rewards/tool_execution_reward_func": 2.0,
798
+ "rewards/visit_tool_reward_func": 0.9642323851585388,
799
+ "step": 45
800
+ },
801
+ {
802
+ "clip_ratio/high_max": 0.0,
803
+ "clip_ratio/high_mean": 0.0,
804
+ "clip_ratio/low_mean": 0.0,
805
+ "clip_ratio/low_min": 0.0,
806
+ "clip_ratio/region_mean": 0.0,
807
+ "epoch": 0.0011139093374660984,
808
+ "grad_norm": 0.026841083756153995,
809
+ "kl": 0.09619140625,
810
+ "learning_rate": 2e-06,
811
+ "loss": -0.0002,
812
+ "step": 46
813
+ },
814
+ {
815
+ "clip_ratio/high_max": 0.0,
816
+ "clip_ratio/high_mean": 0.0,
817
+ "clip_ratio/low_mean": 0.0,
818
+ "clip_ratio/low_min": 0.0,
819
+ "clip_ratio/region_mean": 0.0,
820
+ "epoch": 0.0011381247578457962,
821
+ "grad_norm": 0.026782109601823232,
822
+ "kl": 0.0999755859375,
823
+ "learning_rate": 2e-06,
824
+ "loss": -0.0002,
825
+ "step": 47
826
+ },
827
+ {
828
+ "clip_ratio/high_max": 0.0,
829
+ "clip_ratio/high_mean": 0.0,
830
+ "clip_ratio/low_mean": 0.0,
831
+ "clip_ratio/low_min": 0.0,
832
+ "clip_ratio/region_mean": 0.0,
833
+ "epoch": 0.001162340178225494,
834
+ "grad_norm": 0.02659970469430567,
835
+ "kl": 0.103515625,
836
+ "learning_rate": 2e-06,
837
+ "loss": -0.0002,
838
+ "step": 48
839
+ },
840
+ {
841
+ "clip_ratio/high_max": 0.0,
842
+ "clip_ratio/high_mean": 0.0,
843
+ "clip_ratio/low_mean": 0.0,
844
+ "clip_ratio/low_min": 0.0,
845
+ "clip_ratio/region_mean": 0.0,
846
+ "completions/clipped_ratio": 0.0,
847
+ "completions/max_length": 5239.0,
848
+ "completions/max_terminated_length": 5239.0,
849
+ "completions/mean_length": 3787.1875,
850
+ "completions/mean_terminated_length": 3787.1875,
851
+ "completions/min_length": 2194.0,
852
+ "completions/min_terminated_length": 2194.0,
853
+ "epoch": 0.0011865555986051918,
854
+ "grad_norm": 0.02076190210123295,
855
+ "kl": 0.0938720703125,
856
+ "learning_rate": 2e-06,
857
+ "loss": 0.0001,
858
+ "num_tokens": 4663125.0,
859
+ "reward": 0.7249662280082703,
860
+ "reward_std": 0.16501030325889587,
861
+ "rewards/avg_thinking_length_func": 140.40789794921875,
862
+ "rewards/correct_answer_reward_func": 0.625,
863
+ "rewards/efficient_thinking_reward_func": 0.9519768288322725,
864
+ "rewards/format_reward_func": 1.0,
865
+ "rewards/num_xml_reward_func": 1.8708701133728027,
866
+ "rewards/tool_execution_reward_func": 2.0,
867
+ "rewards/visit_tool_reward_func": 0.9880074858665466,
868
+ "step": 49
869
+ },
870
+ {
871
+ "clip_ratio/high_max": 0.0,
872
+ "clip_ratio/high_mean": 0.0,
873
+ "clip_ratio/low_mean": 0.0,
874
+ "clip_ratio/low_min": 0.0,
875
+ "clip_ratio/region_mean": 0.0,
876
+ "epoch": 0.0012107710189848896,
877
+ "grad_norm": 0.020723077749825757,
878
+ "kl": 0.0958251953125,
879
+ "learning_rate": 2e-06,
880
+ "loss": 0.0001,
881
+ "step": 50
882
+ },
883
+ {
884
+ "clip_ratio/high_max": 0.0,
885
+ "clip_ratio/high_mean": 0.0,
886
+ "clip_ratio/low_mean": 0.0,
887
+ "clip_ratio/low_min": 0.0,
888
+ "clip_ratio/region_mean": 0.0,
889
+ "epoch": 0.0012349864393645874,
890
+ "grad_norm": 0.02054164461884488,
891
+ "kl": 0.09716796875,
892
+ "learning_rate": 2e-06,
893
+ "loss": 0.0001,
894
+ "step": 51
895
+ },
896
+ {
897
+ "clip_ratio/high_max": 0.0,
898
+ "clip_ratio/high_mean": 0.0,
899
+ "clip_ratio/low_mean": 0.0,
900
+ "clip_ratio/low_min": 0.0,
901
+ "clip_ratio/region_mean": 0.0,
902
+ "epoch": 0.0012592018597442853,
903
+ "grad_norm": 0.020335770375883248,
904
+ "kl": 0.0972900390625,
905
+ "learning_rate": 2e-06,
906
+ "loss": 0.0001,
907
+ "step": 52
908
+ },
909
+ {
910
+ "clip_ratio/high_max": 0.0,
911
+ "clip_ratio/high_mean": 0.0,
912
+ "clip_ratio/low_mean": 0.0,
913
+ "clip_ratio/low_min": 0.0,
914
+ "clip_ratio/region_mean": 0.0,
915
+ "completions/clipped_ratio": 0.0,
916
+ "completions/max_length": 5500.0,
917
+ "completions/max_terminated_length": 5500.0,
918
+ "completions/mean_length": 3872.6875,
919
+ "completions/mean_terminated_length": 3872.6875,
920
+ "completions/min_length": 1995.0,
921
+ "completions/min_terminated_length": 1995.0,
922
+ "epoch": 0.001283417280123983,
923
+ "grad_norm": 0.019885369914639652,
924
+ "kl": 0.1060791015625,
925
+ "learning_rate": 2e-06,
926
+ "loss": -0.0001,
927
+ "num_tokens": 4965048.0,
928
+ "reward": 0.8397980332374573,
929
+ "reward_std": 0.24992188811302185,
930
+ "rewards/avg_thinking_length_func": 148.27633666992188,
931
+ "rewards/correct_answer_reward_func": 0.71875,
932
+ "rewards/efficient_thinking_reward_func": 0.9891416230502088,
933
+ "rewards/format_reward_func": 1.0,
934
+ "rewards/num_xml_reward_func": 1.8712050914764404,
935
+ "rewards/tool_execution_reward_func": 2.0,
936
+ "rewards/visit_tool_reward_func": 1.0259138345718384,
937
+ "step": 53
938
+ },
939
+ {
940
+ "clip_ratio/high_max": 0.0,
941
+ "clip_ratio/high_mean": 0.0,
942
+ "clip_ratio/low_mean": 0.0,
943
+ "clip_ratio/low_min": 0.0,
944
+ "clip_ratio/region_mean": 0.0,
945
+ "epoch": 0.0013076327005036806,
946
+ "grad_norm": 0.019647566668717146,
947
+ "kl": 0.1060791015625,
948
+ "learning_rate": 2e-06,
949
+ "loss": -0.0001,
950
+ "step": 54
951
+ },
952
+ {
953
+ "clip_ratio/high_max": 0.0,
954
+ "clip_ratio/high_mean": 0.0,
955
+ "clip_ratio/low_mean": 0.0,
956
+ "clip_ratio/low_min": 0.0,
957
+ "clip_ratio/region_mean": 0.0,
958
+ "epoch": 0.0013318481208833785,
959
+ "grad_norm": 0.01955130786377932,
960
+ "kl": 0.10595703125,
961
+ "learning_rate": 2e-06,
962
+ "loss": -0.0001,
963
+ "step": 55
964
+ },
965
+ {
966
+ "clip_ratio/high_max": 0.0,
967
+ "clip_ratio/high_mean": 0.0,
968
+ "clip_ratio/low_mean": 0.0,
969
+ "clip_ratio/low_min": 0.0,
970
+ "clip_ratio/region_mean": 0.0,
971
+ "epoch": 0.0013560635412630763,
972
+ "grad_norm": 0.01954445177850349,
973
+ "kl": 0.1064453125,
974
+ "learning_rate": 2e-06,
975
+ "loss": -0.0001,
976
+ "step": 56
977
+ },
978
+ {
979
+ "clip_ratio/high_max": 0.0,
980
+ "clip_ratio/high_mean": 0.0,
981
+ "clip_ratio/low_mean": 0.0,
982
+ "clip_ratio/low_min": 0.0,
983
+ "clip_ratio/region_mean": 0.0,
984
+ "completions/clipped_ratio": 0.0,
985
+ "completions/max_length": 5363.0,
986
+ "completions/max_terminated_length": 5363.0,
987
+ "completions/mean_length": 3860.25,
988
+ "completions/mean_terminated_length": 3860.25,
989
+ "completions/min_length": 2462.0,
990
+ "completions/min_terminated_length": 2462.0,
991
+ "epoch": 0.001380278961642774,
992
+ "grad_norm": 0.024884617159235588,
993
+ "kl": 0.093994140625,
994
+ "learning_rate": 2e-06,
995
+ "loss": 0.0007,
996
+ "num_tokens": 5246738.0,
997
+ "reward": 0.6381564140319824,
998
+ "reward_std": 0.36967217922210693,
999
+ "rewards/avg_thinking_length_func": 145.39739990234375,
1000
+ "rewards/correct_answer_reward_func": 0.59375,
1001
+ "rewards/efficient_thinking_reward_func": 0.9513852113551942,
1002
+ "rewards/format_reward_func": 1.0,
1003
+ "rewards/num_xml_reward_func": 1.8717927932739258,
1004
+ "rewards/tool_execution_reward_func": 2.0,
1005
+ "rewards/visit_tool_reward_func": 1.0618340969085693,
1006
+ "step": 57
1007
+ },
1008
+ {
1009
+ "clip_ratio/high_max": 0.0,
1010
+ "clip_ratio/high_mean": 0.0,
1011
+ "clip_ratio/low_mean": 0.0,
1012
+ "clip_ratio/low_min": 0.0,
1013
+ "clip_ratio/region_mean": 0.0,
1014
+ "epoch": 0.0014044943820224719,
1015
+ "grad_norm": 0.02492436560745493,
1016
+ "kl": 0.09375,
1017
+ "learning_rate": 2e-06,
1018
+ "loss": 0.0007,
1019
+ "step": 58
1020
+ },
1021
+ {
1022
+ "clip_ratio/high_max": 0.0,
1023
+ "clip_ratio/high_mean": 0.0,
1024
+ "clip_ratio/low_mean": 0.0,
1025
+ "clip_ratio/low_min": 0.0,
1026
+ "clip_ratio/region_mean": 0.0,
1027
+ "epoch": 0.0014287098024021697,
1028
+ "grad_norm": 0.02501212297767101,
1029
+ "kl": 0.09326171875,
1030
+ "learning_rate": 2e-06,
1031
+ "loss": 0.0007,
1032
+ "step": 59
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
+ "epoch": 0.0014529252227818675,
1041
+ "grad_norm": 0.02474493078785747,
1042
+ "kl": 0.093017578125,
1043
+ "learning_rate": 2e-06,
1044
+ "loss": 0.0007,
1045
+ "step": 60
1046
+ },
1047
+ {
1048
+ "clip_ratio/high_max": 0.0,
1049
+ "clip_ratio/high_mean": 0.0,
1050
+ "clip_ratio/low_mean": 0.0,
1051
+ "clip_ratio/low_min": 0.0,
1052
+ "clip_ratio/region_mean": 0.0,
1053
+ "completions/clipped_ratio": 0.0,
1054
+ "completions/max_length": 6190.0,
1055
+ "completions/max_terminated_length": 6190.0,
1056
+ "completions/mean_length": 4112.90625,
1057
+ "completions/mean_terminated_length": 4112.90625,
1058
+ "completions/min_length": 2503.0,
1059
+ "completions/min_terminated_length": 2503.0,
1060
+ "epoch": 0.0014771406431615653,
1061
+ "grad_norm": 0.02178738104725311,
1062
+ "kl": 0.0994873046875,
1063
+ "learning_rate": 2e-06,
1064
+ "loss": 0.0011,
1065
+ "num_tokens": 5531673.0,
1066
+ "reward": 0.7260021567344666,
1067
+ "reward_std": 0.4103294909000397,
1068
+ "rewards/avg_thinking_length_func": 154.5753936767578,
1069
+ "rewards/correct_answer_reward_func": 0.65625,
1070
+ "rewards/efficient_thinking_reward_func": 0.9408740694270286,
1071
+ "rewards/format_reward_func": 1.0,
1072
+ "rewards/num_xml_reward_func": 1.8727319240570068,
1073
+ "rewards/tool_execution_reward_func": 2.0,
1074
+ "rewards/visit_tool_reward_func": 1.1154723167419434,
1075
+ "step": 61
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
+ "epoch": 0.001501356063541263,
1084
+ "grad_norm": 0.022400760235574114,
1085
+ "kl": 0.09912109375,
1086
+ "learning_rate": 2e-06,
1087
+ "loss": 0.0011,
1088
+ "step": 62
1089
+ },
1090
+ {
1091
+ "clip_ratio/high_max": 0.0,
1092
+ "clip_ratio/high_mean": 0.0,
1093
+ "clip_ratio/low_mean": 0.0,
1094
+ "clip_ratio/low_min": 0.0,
1095
+ "clip_ratio/region_mean": 0.0,
1096
+ "epoch": 0.001525571483920961,
1097
+ "grad_norm": 0.02241024326849516,
1098
+ "kl": 0.0986328125,
1099
+ "learning_rate": 2e-06,
1100
+ "loss": 0.0011,
1101
+ "step": 63
1102
+ },
1103
+ {
1104
+ "clip_ratio/high_max": 0.0,
1105
+ "clip_ratio/high_mean": 0.0,
1106
+ "clip_ratio/low_mean": 0.0,
1107
+ "clip_ratio/low_min": 0.0,
1108
+ "clip_ratio/region_mean": 0.0,
1109
+ "epoch": 0.0015497869043006587,
1110
+ "grad_norm": 0.02284296428778258,
1111
+ "kl": 0.098388671875,
1112
+ "learning_rate": 2e-06,
1113
+ "loss": 0.0011,
1114
+ "step": 64
1115
+ },
1116
+ {
1117
+ "clip_ratio/high_max": 0.0,
1118
+ "clip_ratio/high_mean": 0.0,
1119
+ "clip_ratio/low_mean": 0.0,
1120
+ "clip_ratio/low_min": 0.0,
1121
+ "clip_ratio/region_mean": 0.0,
1122
+ "completions/clipped_ratio": 0.0,
1123
+ "completions/max_length": 6287.0,
1124
+ "completions/max_terminated_length": 6287.0,
1125
+ "completions/mean_length": 4386.625,
1126
+ "completions/mean_terminated_length": 4386.625,
1127
+ "completions/min_length": 3015.0,
1128
+ "completions/min_terminated_length": 3015.0,
1129
+ "epoch": 0.0015740023246803565,
1130
+ "grad_norm": 0.024442904812157877,
1131
+ "kl": 0.096923828125,
1132
+ "learning_rate": 2e-06,
1133
+ "loss": -0.0003,
1134
+ "num_tokens": 5833487.0,
1135
+ "reward": 0.8125966787338257,
1136
+ "reward_std": 0.40657174587249756,
1137
+ "rewards/avg_thinking_length_func": 169.9393310546875,
1138
+ "rewards/correct_answer_reward_func": 0.6875,
1139
+ "rewards/efficient_thinking_reward_func": 0.9483068783516069,
1140
+ "rewards/format_reward_func": 1.0,
1141
+ "rewards/num_xml_reward_func": 1.875028371810913,
1142
+ "rewards/tool_execution_reward_func": 2.0,
1143
+ "rewards/visit_tool_reward_func": 1.0933895111083984,
1144
+ "step": 65
1145
+ },
1146
+ {
1147
+ "clip_ratio/high_max": 0.0,
1148
+ "clip_ratio/high_mean": 0.0,
1149
+ "clip_ratio/low_mean": 0.0,
1150
+ "clip_ratio/low_min": 0.0,
1151
+ "clip_ratio/region_mean": 0.0,
1152
+ "epoch": 0.0015982177450600543,
1153
+ "grad_norm": 0.024525067733314074,
1154
+ "kl": 0.096435546875,
1155
+ "learning_rate": 2e-06,
1156
+ "loss": -0.0003,
1157
+ "step": 66
1158
+ },
1159
+ {
1160
+ "clip_ratio/high_max": 0.0,
1161
+ "clip_ratio/high_mean": 0.0,
1162
+ "clip_ratio/low_mean": 0.0,
1163
+ "clip_ratio/low_min": 0.0,
1164
+ "clip_ratio/region_mean": 0.0,
1165
+ "epoch": 0.0016224331654397521,
1166
+ "grad_norm": 0.024873815112161177,
1167
+ "kl": 0.0960693359375,
1168
+ "learning_rate": 2e-06,
1169
+ "loss": -0.0003,
1170
+ "step": 67
1171
+ },
1172
+ {
1173
+ "clip_ratio/high_max": 0.0,
1174
+ "clip_ratio/high_mean": 0.0,
1175
+ "clip_ratio/low_mean": 0.0,
1176
+ "clip_ratio/low_min": 0.0,
1177
+ "clip_ratio/region_mean": 0.0,
1178
+ "epoch": 0.00164664858581945,
1179
+ "grad_norm": 0.025222911405096846,
1180
+ "kl": 0.0955810546875,
1181
+ "learning_rate": 2e-06,
1182
+ "loss": -0.0003,
1183
+ "step": 68
1184
+ },
1185
+ {
1186
+ "clip_ratio/high_max": 0.0,
1187
+ "clip_ratio/high_mean": 0.0,
1188
+ "clip_ratio/low_mean": 0.0,
1189
+ "clip_ratio/low_min": 0.0,
1190
+ "clip_ratio/region_mean": 0.0,
1191
+ "completions/clipped_ratio": 0.0,
1192
+ "completions/max_length": 7582.0,
1193
+ "completions/max_terminated_length": 7582.0,
1194
+ "completions/mean_length": 4336.375,
1195
+ "completions/mean_terminated_length": 4336.375,
1196
+ "completions/min_length": 2926.0,
1197
+ "completions/min_terminated_length": 2926.0,
1198
+ "epoch": 0.0016708640061991475,
1199
+ "grad_norm": 0.017478901741330728,
1200
+ "kl": 0.0911865234375,
1201
+ "learning_rate": 2e-06,
1202
+ "loss": 0.0019,
1203
+ "num_tokens": 6132990.0,
1204
+ "reward": 0.7490283846855164,
1205
+ "reward_std": 0.15198753774166107,
1206
+ "rewards/avg_thinking_length_func": 179.77679443359375,
1207
+ "rewards/correct_answer_reward_func": 0.65625,
1208
+ "rewards/efficient_thinking_reward_func": 0.938588221110238,
1209
+ "rewards/format_reward_func": 1.0,
1210
+ "rewards/num_xml_reward_func": 1.9357295036315918,
1211
+ "rewards/tool_execution_reward_func": 2.0,
1212
+ "rewards/visit_tool_reward_func": 0.9721688032150269,
1213
+ "step": 69
1214
+ },
1215
+ {
1216
+ "clip_ratio/high_max": 0.0,
1217
+ "clip_ratio/high_mean": 0.0,
1218
+ "clip_ratio/low_mean": 0.0,
1219
+ "clip_ratio/low_min": 0.0,
1220
+ "clip_ratio/region_mean": 0.0,
1221
+ "epoch": 0.0016950794265788453,
1222
+ "grad_norm": 0.018875622948253925,
1223
+ "kl": 0.0909423828125,
1224
+ "learning_rate": 2e-06,
1225
+ "loss": 0.0019,
1226
+ "step": 70
1227
+ },
1228
+ {
1229
+ "clip_ratio/high_max": 0.0,
1230
+ "clip_ratio/high_mean": 0.0,
1231
+ "clip_ratio/low_mean": 0.0,
1232
+ "clip_ratio/low_min": 0.0,
1233
+ "clip_ratio/region_mean": 0.0,
1234
+ "epoch": 0.0017192948469585431,
1235
+ "grad_norm": 0.020484617082338733,
1236
+ "kl": 0.0911865234375,
1237
+ "learning_rate": 2e-06,
1238
+ "loss": 0.0019,
1239
+ "step": 71
1240
+ },
1241
+ {
1242
+ "clip_ratio/high_max": 0.0,
1243
+ "clip_ratio/high_mean": 0.0,
1244
+ "clip_ratio/low_mean": 0.0,
1245
+ "clip_ratio/low_min": 0.0,
1246
+ "clip_ratio/region_mean": 0.0,
1247
+ "epoch": 0.001743510267338241,
1248
+ "grad_norm": 0.021532205957360162,
1249
+ "kl": 0.0933837890625,
1250
+ "learning_rate": 2e-06,
1251
+ "loss": 0.0019,
1252
+ "step": 72
1253
+ },
1254
+ {
1255
+ "clip_ratio/high_max": 0.0,
1256
+ "clip_ratio/high_mean": 0.0,
1257
+ "clip_ratio/low_mean": 0.0,
1258
+ "clip_ratio/low_min": 0.0,
1259
+ "clip_ratio/region_mean": 0.0,
1260
+ "completions/clipped_ratio": 0.0,
1261
+ "completions/max_length": 6065.0,
1262
+ "completions/max_terminated_length": 6065.0,
1263
+ "completions/mean_length": 4207.40625,
1264
+ "completions/mean_terminated_length": 4207.40625,
1265
+ "completions/min_length": 2648.0,
1266
+ "completions/min_terminated_length": 2648.0,
1267
+ "epoch": 0.0017677256877179387,
1268
+ "grad_norm": 0.02249842465908536,
1269
+ "kl": 0.1011962890625,
1270
+ "learning_rate": 2e-06,
1271
+ "loss": 0.0006,
1272
+ "num_tokens": 6415549.0,
1273
+ "reward": 0.7511920928955078,
1274
+ "reward_std": 0.23255480825901031,
1275
+ "rewards/avg_thinking_length_func": 180.5469207763672,
1276
+ "rewards/correct_answer_reward_func": 0.65625,
1277
+ "rewards/efficient_thinking_reward_func": 0.9317484663097895,
1278
+ "rewards/format_reward_func": 1.0,
1279
+ "rewards/num_xml_reward_func": 1.9344795942306519,
1280
+ "rewards/tool_execution_reward_func": 1.990625023841858,
1281
+ "rewards/visit_tool_reward_func": 0.9795504808425903,
1282
+ "step": 73
1283
+ },
1284
+ {
1285
+ "clip_ratio/high_max": 0.0,
1286
+ "clip_ratio/high_mean": 0.0,
1287
+ "clip_ratio/low_mean": 0.0,
1288
+ "clip_ratio/low_min": 0.0,
1289
+ "clip_ratio/region_mean": 0.0,
1290
+ "epoch": 0.0017919411080976365,
1291
+ "grad_norm": 0.05340931146356945,
1292
+ "kl": 0.13232421875,
1293
+ "learning_rate": 2e-06,
1294
+ "loss": 0.0007,
1295
+ "step": 74
1296
+ },
1297
+ {
1298
+ "clip_ratio/high_max": 0.0,
1299
+ "clip_ratio/high_mean": 0.0,
1300
+ "clip_ratio/low_mean": 0.0,
1301
+ "clip_ratio/low_min": 0.0,
1302
+ "clip_ratio/region_mean": 0.0,
1303
+ "epoch": 0.0018161565284773344,
1304
+ "grad_norm": 0.02675181315227704,
1305
+ "kl": 0.1058349609375,
1306
+ "learning_rate": 2e-06,
1307
+ "loss": 0.0006,
1308
+ "step": 75
1309
+ },
1310
+ {
1311
+ "clip_ratio/high_max": 0.0,
1312
+ "clip_ratio/high_mean": 0.0,
1313
+ "clip_ratio/low_mean": 0.0,
1314
+ "clip_ratio/low_min": 0.0,
1315
+ "clip_ratio/region_mean": 0.0,
1316
+ "epoch": 0.0018403719488570322,
1317
+ "grad_norm": 0.02697259723804417,
1318
+ "kl": 0.091064453125,
1319
+ "learning_rate": 2e-06,
1320
+ "loss": 0.0006,
1321
+ "step": 76
1322
+ },
1323
+ {
1324
+ "clip_ratio/high_max": 0.0,
1325
+ "clip_ratio/high_mean": 0.0,
1326
+ "clip_ratio/low_mean": 0.0,
1327
+ "clip_ratio/low_min": 0.0,
1328
+ "clip_ratio/region_mean": 0.0,
1329
+ "completions/clipped_ratio": 0.0,
1330
+ "completions/max_length": 6482.0,
1331
+ "completions/max_terminated_length": 6482.0,
1332
+ "completions/mean_length": 4148.40625,
1333
+ "completions/mean_terminated_length": 4148.40625,
1334
+ "completions/min_length": 2778.0,
1335
+ "completions/min_terminated_length": 2778.0,
1336
+ "epoch": 0.00186458736923673,
1337
+ "grad_norm": 0.03330572791868595,
1338
+ "kl": 0.096923828125,
1339
+ "learning_rate": 2e-06,
1340
+ "loss": 0.0014,
1341
+ "num_tokens": 6703449.0,
1342
+ "reward": 0.6695541739463806,
1343
+ "reward_std": 0.3436514735221863,
1344
+ "rewards/avg_thinking_length_func": 184.87484741210938,
1345
+ "rewards/correct_answer_reward_func": 0.59375,
1346
+ "rewards/efficient_thinking_reward_func": 0.9242631827661938,
1347
+ "rewards/format_reward_func": 1.0,
1348
+ "rewards/num_xml_reward_func": 1.9313738346099854,
1349
+ "rewards/tool_execution_reward_func": 2.0,
1350
+ "rewards/visit_tool_reward_func": 0.9495425224304199,
1351
+ "step": 77
1352
+ },
1353
+ {
1354
+ "clip_ratio/high_max": 0.0,
1355
+ "clip_ratio/high_mean": 0.0,
1356
+ "clip_ratio/low_mean": 0.0,
1357
+ "clip_ratio/low_min": 0.0,
1358
+ "clip_ratio/region_mean": 0.0,
1359
+ "epoch": 0.0018888027896164278,
1360
+ "grad_norm": 0.03702086074093763,
1361
+ "kl": 0.0970458984375,
1362
+ "learning_rate": 2e-06,
1363
+ "loss": 0.0014,
1364
+ "step": 78
1365
+ },
1366
+ {
1367
+ "clip_ratio/high_max": 0.0,
1368
+ "clip_ratio/high_mean": 0.0,
1369
+ "clip_ratio/low_mean": 0.0,
1370
+ "clip_ratio/low_min": 0.0,
1371
+ "clip_ratio/region_mean": 0.0,
1372
+ "epoch": 0.0019130182099961256,
1373
+ "grad_norm": 0.039763093996501166,
1374
+ "kl": 0.1046142578125,
1375
+ "learning_rate": 2e-06,
1376
+ "loss": 0.0015,
1377
+ "step": 79
1378
+ },
1379
+ {
1380
+ "clip_ratio/high_max": 0.0,
1381
+ "clip_ratio/high_mean": 0.0,
1382
+ "clip_ratio/low_mean": 0.0,
1383
+ "clip_ratio/low_min": 0.0,
1384
+ "clip_ratio/region_mean": 0.0,
1385
+ "epoch": 0.0019372336303758234,
1386
+ "grad_norm": 0.04490759555701776,
1387
+ "kl": 0.1263427734375,
1388
+ "learning_rate": 2e-06,
1389
+ "loss": 0.0015,
1390
+ "step": 80
1391
+ }
1392
+ ],
1393
+ "logging_steps": 1,
1394
+ "max_steps": 640,
1395
+ "num_input_tokens_seen": 6703449,
1396
+ "num_train_epochs": 1,
1397
+ "save_steps": 20,
1398
+ "stateful_callbacks": {
1399
+ "TrainerControl": {
1400
+ "args": {
1401
+ "should_epoch_stop": false,
1402
+ "should_evaluate": false,
1403
+ "should_log": false,
1404
+ "should_save": true,
1405
+ "should_training_stop": false
1406
+ },
1407
+ "attributes": {}
1408
+ }
1409
+ },
1410
+ "total_flos": 0.0,
1411
+ "train_batch_size": 2,
1412
+ "trial_name": null,
1413
+ "trial_params": null
1414
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11bcd76e4cac994c7f4a4f0f72f38922a2e0592872e99e4bfd16fb340f40dfb4
3
+ size 8465
vocab.json ADDED
The diff for this file is too large to render. See raw diff