Saad4web commited on
Commit
29095e4
·
verified ·
1 Parent(s): 20e68c2

Initial upload of FunctionGemma Director V1

Browse files
Files changed (48) hide show
  1. .gitattributes +3 -0
  2. README.md +58 -0
  3. added_tokens.json +4 -0
  4. all_results.json +8 -0
  5. chat_template.jinja +279 -0
  6. checkpoint-50/added_tokens.json +4 -0
  7. checkpoint-50/chat_template.jinja +279 -0
  8. checkpoint-50/config.json +54 -0
  9. checkpoint-50/generation_config.json +15 -0
  10. checkpoint-50/model.safetensors +3 -0
  11. checkpoint-50/optimizer.pt +3 -0
  12. checkpoint-50/rng_state.pth +3 -0
  13. checkpoint-50/scheduler.pt +3 -0
  14. checkpoint-50/special_tokens_map.json +34 -0
  15. checkpoint-50/tokenizer.json +3 -0
  16. checkpoint-50/tokenizer.model +3 -0
  17. checkpoint-50/tokenizer_config.json +0 -0
  18. checkpoint-50/trainer_state.json +104 -0
  19. checkpoint-50/training_args.bin +3 -0
  20. checkpoint-66/added_tokens.json +4 -0
  21. checkpoint-66/chat_template.jinja +279 -0
  22. checkpoint-66/config.json +54 -0
  23. checkpoint-66/generation_config.json +15 -0
  24. checkpoint-66/model.safetensors +3 -0
  25. checkpoint-66/optimizer.pt +3 -0
  26. checkpoint-66/rng_state.pth +3 -0
  27. checkpoint-66/scheduler.pt +3 -0
  28. checkpoint-66/special_tokens_map.json +34 -0
  29. checkpoint-66/tokenizer.json +3 -0
  30. checkpoint-66/tokenizer.model +3 -0
  31. checkpoint-66/tokenizer_config.json +0 -0
  32. checkpoint-66/trainer_state.json +125 -0
  33. checkpoint-66/training_args.bin +3 -0
  34. config.json +54 -0
  35. generation_config.json +15 -0
  36. model.safetensors +3 -0
  37. runs/Jan02_13-47-02_074ad1548a3d/events.out.tfevents.1767361678.074ad1548a3d.3005.0 +3 -0
  38. runs/Jan02_13-50-09_074ad1548a3d/events.out.tfevents.1767361816.074ad1548a3d.4055.0 +3 -0
  39. runs/Jan02_13-53-13_074ad1548a3d/events.out.tfevents.1767362039.074ad1548a3d.4947.0 +3 -0
  40. special_tokens_map.json +34 -0
  41. tokenizer.json +3 -0
  42. tokenizer.model +3 -0
  43. tokenizer_config.json +0 -0
  44. train_results.json +8 -0
  45. trainer_log.jsonl +14 -0
  46. trainer_state.json +134 -0
  47. training_args.bin +3 -0
  48. training_loss.png +0 -0
.gitattributes CHANGED
@@ -33,3 +33,6 @@ 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
+ checkpoint-50/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ checkpoint-66/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: other
4
+ base_model: google/functiongemma-270m-it
5
+ tags:
6
+ - llama-factory
7
+ - full
8
+ - generated_from_trainer
9
+ model-index:
10
+ - name: FunctionGemma_Director_V1
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # FunctionGemma_Director_V1
18
+
19
+ This model is a fine-tuned version of [google/functiongemma-270m-it](https://huggingface.co/google/functiongemma-270m-it) on the game_director dataset.
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 5e-05
39
+ - train_batch_size: 2
40
+ - eval_batch_size: 8
41
+ - seed: 42
42
+ - gradient_accumulation_steps: 8
43
+ - total_train_batch_size: 16
44
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
45
+ - lr_scheduler_type: cosine
46
+ - lr_scheduler_warmup_ratio: 0.1
47
+ - num_epochs: 3.0
48
+
49
+ ### Training results
50
+
51
+
52
+
53
+ ### Framework versions
54
+
55
+ - Transformers 4.57.1
56
+ - Pytorch 2.9.0+cu126
57
+ - Datasets 4.0.0
58
+ - Tokenizers 0.22.1
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<end_of_image>": 262145,
3
+ "<image_soft_token>": 262144
4
+ }
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 411612762796032.0,
4
+ "train_loss": 0.12000229429792274,
5
+ "train_runtime": 714.9914,
6
+ "train_samples_per_second": 1.435,
7
+ "train_steps_per_second": 0.092
8
+ }
chat_template.jinja ADDED
@@ -0,0 +1,279 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- macro format_parameters(properties, required) -%}
2
+ {%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%}
3
+ {%- set ns = namespace(found_first=false) -%}
4
+ {%- for key, value in properties | dictsort -%}
5
+ {%- if key not in standard_keys -%}
6
+ {%- if ns.found_first %},{% endif -%}
7
+ {%- set ns.found_first = true -%}
8
+ {{- key }}:{description:<escape>{{ value['description'] }}<escape>
9
+ {%- if value['type'] | upper == 'STRING' -%}
10
+ {%- if value['enum'] -%}
11
+ ,enum:{{ format_argument(value['enum']) }}
12
+ {%- endif -%}
13
+ {%- elif value['type'] | upper == 'OBJECT' -%}
14
+ ,properties:{
15
+ {%- if value['properties'] is defined and value['properties'] is mapping -%}
16
+ {{- format_parameters(value['properties'], value['required'] | default([])) -}}
17
+ {%- elif value is mapping -%}
18
+ {{- format_parameters(value, value['required'] | default([])) -}}
19
+ {%- endif -%}
20
+ }
21
+ {%- if value['required'] -%}
22
+ ,required:[
23
+ {%- for item in value['required'] | default([]) -%}
24
+ <escape>{{- item -}}<escape>
25
+ {%- if not loop.last %},{% endif -%}
26
+ {%- endfor -%}
27
+ ]
28
+ {%- endif -%}
29
+ {%- elif value['type'] | upper == 'ARRAY' -%}
30
+ {%- if value['items'] is mapping and value['items'] -%}
31
+ ,items:{
32
+ {%- set ns_items = namespace(found_first=false) -%}
33
+ {%- for item_key, item_value in value['items'] | dictsort -%}
34
+ {%- if item_value is not none -%}
35
+ {%- if ns_items.found_first %},{% endif -%}
36
+ {%- set ns_items.found_first = true -%}
37
+ {%- if item_key == 'properties' -%}
38
+ properties:{
39
+ {%- if item_value is mapping -%}
40
+ {{- format_parameters(item_value, value['items']['required'] | default([])) -}}
41
+ {%- endif -%}
42
+ }
43
+ {%- elif item_key == 'required' -%}
44
+ required:[
45
+ {%- for req_item in item_value -%}
46
+ <escape>{{- req_item -}}<escape>
47
+ {%- if not loop.last %},{% endif -%}
48
+ {%- endfor -%}
49
+ ]
50
+ {%- elif item_key == 'type' -%}
51
+ {%- if item_value is string -%}
52
+ type:{{ format_argument(item_value | upper) }}
53
+ {%- else -%}
54
+ type:{{ format_argument(item_value | map('upper') | list) }}
55
+ {%- endif -%}
56
+ {%- else -%}
57
+ {{ item_key }}:{{ format_argument(item_value) }}
58
+ {%- endif -%}
59
+ {%- endif -%}
60
+ {%- endfor -%}
61
+ }
62
+ {%- endif -%}
63
+ {%- endif -%}
64
+ ,type:<escape>{{ value['type'] | upper }}<escape>}
65
+ {%- endif -%}
66
+ {%- endfor -%}
67
+ {%- endmacro -%}
68
+ {% macro format_function_declaration(tool_data) -%}
69
+ declaration:{{- tool_data['function']['name'] -}}
70
+ {description:<escape>{{- tool_data['function']['description'] -}}<escape>
71
+ {%- set params = tool_data['function']['parameters'] -%}
72
+ {%- if params -%}
73
+ ,parameters:{
74
+ {%- if params['properties'] -%}
75
+ properties:{ {{- format_parameters(params['properties'], params['required']) -}} },
76
+ {%- endif -%}
77
+ {%- if params['required'] -%}
78
+ required:[
79
+ {%- for item in params['required'] -%}
80
+ <escape>{{- item -}}<escape>
81
+ {{- ',' if not loop.last -}}
82
+ {%- endfor -%}
83
+ ],
84
+ {%- endif -%}
85
+ {%- if params['type'] -%}
86
+ type:<escape>{{- params['type'] | upper -}}<escape>}
87
+ {%- endif -%}
88
+ {%- endif -%}
89
+ }
90
+ {%- endmacro -%}
91
+ {% macro format_argument(argument, escape_keys=True) -%}
92
+ {%- if argument is string -%}
93
+ {{- '<escape>' + argument + '<escape>' -}}
94
+ {%- elif argument is boolean -%}
95
+ {%- if argument -%}
96
+ {{- 'true' -}}
97
+ {%- else -%}
98
+ {{- 'false' -}}
99
+ {%- endif -%}
100
+ {%- elif argument is mapping -%}
101
+ {{- '{' -}}
102
+ {%- set ns = namespace(found_first=false) -%}
103
+ {%- for key, value in argument | dictsort -%}
104
+ {%- if ns.found_first %},{% endif -%}
105
+ {%- set ns.found_first = true -%}
106
+ {%- if escape_keys -%}
107
+ {{- '<escape>' + key + '<escape>' -}}
108
+ {%- else -%}
109
+ {{- key -}}
110
+ {%- endif -%}
111
+ :{{- format_argument(value, escape_keys=escape_keys) -}}
112
+ {%- endfor -%}
113
+ {{- '}' -}}
114
+ {%- elif argument is sequence -%}
115
+ {{- '[' -}}
116
+ {%- for item in argument -%}
117
+ {{- format_argument(item, escape_keys=escape_keys) -}}
118
+ {%- if not loop.last %},{% endif -%}
119
+ {%- endfor -%}
120
+ {{- ']' -}}
121
+ {%- else -%}
122
+ {{- argument -}}
123
+ {%- endif -%}
124
+ {%- endmacro -%}
125
+ {{ bos_token }}
126
+ {%- set ns = namespace(prev_message_type=None) -%}
127
+ {#- Tool Declarations -#}
128
+ {%- set loop_messages = messages -%}
129
+ {%- if tools or messages[0]['role'] == 'system' or messages[0]['role'] == 'developer' -%}
130
+ {{- '<start_of_turn>developer\n' -}}
131
+ {%- if messages[0]['role'] == 'system' or messages[0]['role'] == 'developer' -%}
132
+ {%- if messages[0]['content'] is string -%}
133
+ {{- messages[0]['content'] | trim -}}
134
+ {%- elif messages[0]['content'] is sequence -%}
135
+ {%- for item in messages[0]['content'] -%}
136
+ {%- if item['type'] == 'text' -%}
137
+ {{- item['text'] | trim -}}
138
+ {%- endif -%}
139
+ {%- endfor -%}
140
+ {%- endif -%}
141
+ {%- set loop_messages = messages[1:] -%}
142
+ {%- endif -%}
143
+ {%- if tools -%}
144
+ {%- for tool in tools %}
145
+ {{- '<start_function_declaration>' -}}
146
+ {{- format_function_declaration(tool) | trim }}
147
+ {{- '<end_function_declaration>' -}}
148
+ {%- endfor %}
149
+ {%- endif -%}
150
+ {{- '<end_of_turn>\n' }}
151
+ {%- endif %}
152
+ {#- Loop through messages. -#}
153
+ {%- for message in loop_messages -%}
154
+ {%- if (message['role'] == 'assistant') -%}
155
+ {#- Rename "assistant" to "model". -#}
156
+ {%- set role = "model" -%}
157
+ {%- else -%}
158
+ {%- set role = message['role'] -%}
159
+ {%- endif -%}
160
+ {%- if role != 'tool' -%}
161
+ {%- if ns.prev_message_type != 'tool_response' -%}
162
+ {{- '<start_of_turn>' + role + '\n' }}
163
+ {%- endif -%}
164
+ {%- set ns.prev_message_type = None -%}
165
+ {%- if 'content' in message and message['content'] is not none -%}
166
+ {%- if message['content'] is string -%}
167
+ {{ message['content'] | trim }}
168
+ {%- elif message['content'] is sequence -%}
169
+ {%- for item in message['content'] -%}
170
+ {%- if item['type'] == 'image' -%}
171
+ {{ '<start_of_image>' }}
172
+ {%- elif item['type'] == 'text' -%}
173
+ {{ item['text'] | trim }}
174
+ {%- endif -%}
175
+ {%- endfor -%}
176
+ {%- else -%}
177
+ {{ raise_exception("Invalid content type in user/assistant message") }}
178
+ {%- endif -%}
179
+ {%- set ns.prev_message_type = 'content' -%}
180
+ {%- endif -%}
181
+ {%- if 'tool_calls' in message and message['tool_calls'] and message['tool_calls'] is iterable -%}
182
+ {#- Tool Calls -#}
183
+ {%- for tool_call in message['tool_calls'] -%}
184
+ {% set function = tool_call['function'] %}
185
+ {{- '<start_function_call>call:' + function['name'] + '{' -}}
186
+ {%- if 'arguments' in function -%}
187
+ {%- if function['arguments'] is mapping -%}
188
+ {%- set ns = namespace(found_first=false) -%}
189
+ {%- for key, value in function['arguments'] | dictsort -%}
190
+ {%- if ns.found_first %},{% endif -%}
191
+ {%- set ns.found_first = true -%}
192
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
193
+ {%- endfor -%}
194
+ {%- elif function['arguments'] is string -%}
195
+ {# This handles string-JSON, just in case #}
196
+ {{ function['arguments'] }}
197
+ {%- endif %}
198
+ {%- endif -%}
199
+ {{- '}<end_function_call>' -}}
200
+ {%- endfor -%}
201
+ {%- if loop.last -%}
202
+ {{ '<start_function_response>' }}
203
+ {%- endif -%}
204
+ {%- set ns.prev_message_type = 'tool_call' -%}
205
+ {%- endif -%}
206
+ {%- else -%}
207
+ {#- Tool Responses -#}
208
+ {%- if 'content' in message and message['content'] -%}
209
+ {%- if message['content'] is mapping -%}
210
+ {%- if 'name' in message['content'] and 'response' in message['content'] -%}
211
+ {{ '<start_function_response>response:' + message['content']['name'] | trim + '{' }}
212
+ {%- set response_ns = namespace(found_first=false) -%}
213
+ {%- for key, value in message['content']['response'] | dictsort -%}
214
+ {%- if response_ns.found_first %},{% endif -%}
215
+ {%- set response_ns.found_first = true -%}
216
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
217
+ {%- endfor -%}
218
+ {{- '}<end_function_response>' -}}
219
+ {%- elif 'name' in message -%}
220
+ {{ '<start_function_response>response:' + message['name'] | trim + '{' }}
221
+ {%- set response_ns = namespace(found_first=false) -%}
222
+ {%- for key, value in message['content'] | dictsort -%}
223
+ {%- if response_ns.found_first %},{% endif -%}
224
+ {%- set response_ns.found_first = true -%}
225
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
226
+ {%- endfor -%}
227
+ {{- '}<end_function_response>' -}}
228
+ {%- else -%}
229
+ {{ raise_exception("Invalid tool response mapping: must contain 'name' and 'response' keys, or 'name' must be in the message.") }}
230
+ {%- endif -%}
231
+ {%- elif message['content'] is string -%}
232
+ {%- if 'name' in message -%}
233
+ {{ '<start_function_response>response:' + message['name'] | trim + '{value:' + format_argument(message['content'], escape_keys=False) + '}<end_function_response>' }}
234
+ {%- else -%}
235
+ {{ raise_exception("Invalid tool response: 'name' must be provided.") }}
236
+ {%- endif -%}
237
+ {%- elif message['content'] is sequence -%}
238
+ {%- for item in message['content'] -%}
239
+ {%- if item is mapping -%}
240
+ {%- if 'name' in item and 'response' in item -%}
241
+ {{ '<start_function_response>response:' + item['name'] | trim + '{' }}
242
+ {%- set response_ns = namespace(found_first=false) -%}
243
+ {%- for key, value in item['response'] | dictsort -%}
244
+ {%- if response_ns.found_first %},{% endif -%}
245
+ {%- set response_ns.found_first = true -%}
246
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
247
+ {%- endfor -%}
248
+ {{- '}<end_function_response>' -}}
249
+ {%- elif 'name' in message -%}
250
+ {{ '<start_function_response>response:' + message['name'] | trim + '{' }}
251
+ {%- set response_ns = namespace(found_first=false) -%}
252
+ {%- for key, value in item | dictsort -%}
253
+ {%- if response_ns.found_first %},{% endif -%}
254
+ {%- set response_ns.found_first = true -%}
255
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
256
+ {%- endfor -%}
257
+ {{- '}<end_function_response>' -}}
258
+ {%- else -%}
259
+ {{ raise_exception("Invalid tool response mapping: must contain 'name' and 'response' keys, or 'name' must be in the message.") }}
260
+ {%- endif -%}
261
+ {%- else -%}
262
+ {{ raise_exception("Invalid tool response message: multiple responses must all be mappings") }}
263
+ {%- endif -%}
264
+ {%- endfor -%}
265
+ {%- else -%}
266
+ {{ raise_exception("Invalid content type in tool message: must be mapping, sequence of mappings, or string.") }}
267
+ {%- endif -%}
268
+ {%- endif -%}
269
+ {%- set ns.prev_message_type = 'tool_response' -%}
270
+ {%- endif -%}
271
+ {%- if ns.prev_message_type not in ['tool_call', 'tool_response'] -%}
272
+ {{ '<end_of_turn>\n' }}
273
+ {%- endif -%}
274
+ {%- endfor -%}
275
+ {%- if add_generation_prompt -%}
276
+ {%- if ns.prev_message_type != 'tool_response' -%}
277
+ {{- '<start_of_turn>model\n' -}}
278
+ {%- endif -%}
279
+ {%- endif -%}
checkpoint-50/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<end_of_image>": 262145,
3
+ "<image_soft_token>": 262144
4
+ }
checkpoint-50/chat_template.jinja ADDED
@@ -0,0 +1,279 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- macro format_parameters(properties, required) -%}
2
+ {%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%}
3
+ {%- set ns = namespace(found_first=false) -%}
4
+ {%- for key, value in properties | dictsort -%}
5
+ {%- if key not in standard_keys -%}
6
+ {%- if ns.found_first %},{% endif -%}
7
+ {%- set ns.found_first = true -%}
8
+ {{- key }}:{description:<escape>{{ value['description'] }}<escape>
9
+ {%- if value['type'] | upper == 'STRING' -%}
10
+ {%- if value['enum'] -%}
11
+ ,enum:{{ format_argument(value['enum']) }}
12
+ {%- endif -%}
13
+ {%- elif value['type'] | upper == 'OBJECT' -%}
14
+ ,properties:{
15
+ {%- if value['properties'] is defined and value['properties'] is mapping -%}
16
+ {{- format_parameters(value['properties'], value['required'] | default([])) -}}
17
+ {%- elif value is mapping -%}
18
+ {{- format_parameters(value, value['required'] | default([])) -}}
19
+ {%- endif -%}
20
+ }
21
+ {%- if value['required'] -%}
22
+ ,required:[
23
+ {%- for item in value['required'] | default([]) -%}
24
+ <escape>{{- item -}}<escape>
25
+ {%- if not loop.last %},{% endif -%}
26
+ {%- endfor -%}
27
+ ]
28
+ {%- endif -%}
29
+ {%- elif value['type'] | upper == 'ARRAY' -%}
30
+ {%- if value['items'] is mapping and value['items'] -%}
31
+ ,items:{
32
+ {%- set ns_items = namespace(found_first=false) -%}
33
+ {%- for item_key, item_value in value['items'] | dictsort -%}
34
+ {%- if item_value is not none -%}
35
+ {%- if ns_items.found_first %},{% endif -%}
36
+ {%- set ns_items.found_first = true -%}
37
+ {%- if item_key == 'properties' -%}
38
+ properties:{
39
+ {%- if item_value is mapping -%}
40
+ {{- format_parameters(item_value, value['items']['required'] | default([])) -}}
41
+ {%- endif -%}
42
+ }
43
+ {%- elif item_key == 'required' -%}
44
+ required:[
45
+ {%- for req_item in item_value -%}
46
+ <escape>{{- req_item -}}<escape>
47
+ {%- if not loop.last %},{% endif -%}
48
+ {%- endfor -%}
49
+ ]
50
+ {%- elif item_key == 'type' -%}
51
+ {%- if item_value is string -%}
52
+ type:{{ format_argument(item_value | upper) }}
53
+ {%- else -%}
54
+ type:{{ format_argument(item_value | map('upper') | list) }}
55
+ {%- endif -%}
56
+ {%- else -%}
57
+ {{ item_key }}:{{ format_argument(item_value) }}
58
+ {%- endif -%}
59
+ {%- endif -%}
60
+ {%- endfor -%}
61
+ }
62
+ {%- endif -%}
63
+ {%- endif -%}
64
+ ,type:<escape>{{ value['type'] | upper }}<escape>}
65
+ {%- endif -%}
66
+ {%- endfor -%}
67
+ {%- endmacro -%}
68
+ {% macro format_function_declaration(tool_data) -%}
69
+ declaration:{{- tool_data['function']['name'] -}}
70
+ {description:<escape>{{- tool_data['function']['description'] -}}<escape>
71
+ {%- set params = tool_data['function']['parameters'] -%}
72
+ {%- if params -%}
73
+ ,parameters:{
74
+ {%- if params['properties'] -%}
75
+ properties:{ {{- format_parameters(params['properties'], params['required']) -}} },
76
+ {%- endif -%}
77
+ {%- if params['required'] -%}
78
+ required:[
79
+ {%- for item in params['required'] -%}
80
+ <escape>{{- item -}}<escape>
81
+ {{- ',' if not loop.last -}}
82
+ {%- endfor -%}
83
+ ],
84
+ {%- endif -%}
85
+ {%- if params['type'] -%}
86
+ type:<escape>{{- params['type'] | upper -}}<escape>}
87
+ {%- endif -%}
88
+ {%- endif -%}
89
+ }
90
+ {%- endmacro -%}
91
+ {% macro format_argument(argument, escape_keys=True) -%}
92
+ {%- if argument is string -%}
93
+ {{- '<escape>' + argument + '<escape>' -}}
94
+ {%- elif argument is boolean -%}
95
+ {%- if argument -%}
96
+ {{- 'true' -}}
97
+ {%- else -%}
98
+ {{- 'false' -}}
99
+ {%- endif -%}
100
+ {%- elif argument is mapping -%}
101
+ {{- '{' -}}
102
+ {%- set ns = namespace(found_first=false) -%}
103
+ {%- for key, value in argument | dictsort -%}
104
+ {%- if ns.found_first %},{% endif -%}
105
+ {%- set ns.found_first = true -%}
106
+ {%- if escape_keys -%}
107
+ {{- '<escape>' + key + '<escape>' -}}
108
+ {%- else -%}
109
+ {{- key -}}
110
+ {%- endif -%}
111
+ :{{- format_argument(value, escape_keys=escape_keys) -}}
112
+ {%- endfor -%}
113
+ {{- '}' -}}
114
+ {%- elif argument is sequence -%}
115
+ {{- '[' -}}
116
+ {%- for item in argument -%}
117
+ {{- format_argument(item, escape_keys=escape_keys) -}}
118
+ {%- if not loop.last %},{% endif -%}
119
+ {%- endfor -%}
120
+ {{- ']' -}}
121
+ {%- else -%}
122
+ {{- argument -}}
123
+ {%- endif -%}
124
+ {%- endmacro -%}
125
+ {{ bos_token }}
126
+ {%- set ns = namespace(prev_message_type=None) -%}
127
+ {#- Tool Declarations -#}
128
+ {%- set loop_messages = messages -%}
129
+ {%- if tools or messages[0]['role'] == 'system' or messages[0]['role'] == 'developer' -%}
130
+ {{- '<start_of_turn>developer\n' -}}
131
+ {%- if messages[0]['role'] == 'system' or messages[0]['role'] == 'developer' -%}
132
+ {%- if messages[0]['content'] is string -%}
133
+ {{- messages[0]['content'] | trim -}}
134
+ {%- elif messages[0]['content'] is sequence -%}
135
+ {%- for item in messages[0]['content'] -%}
136
+ {%- if item['type'] == 'text' -%}
137
+ {{- item['text'] | trim -}}
138
+ {%- endif -%}
139
+ {%- endfor -%}
140
+ {%- endif -%}
141
+ {%- set loop_messages = messages[1:] -%}
142
+ {%- endif -%}
143
+ {%- if tools -%}
144
+ {%- for tool in tools %}
145
+ {{- '<start_function_declaration>' -}}
146
+ {{- format_function_declaration(tool) | trim }}
147
+ {{- '<end_function_declaration>' -}}
148
+ {%- endfor %}
149
+ {%- endif -%}
150
+ {{- '<end_of_turn>\n' }}
151
+ {%- endif %}
152
+ {#- Loop through messages. -#}
153
+ {%- for message in loop_messages -%}
154
+ {%- if (message['role'] == 'assistant') -%}
155
+ {#- Rename "assistant" to "model". -#}
156
+ {%- set role = "model" -%}
157
+ {%- else -%}
158
+ {%- set role = message['role'] -%}
159
+ {%- endif -%}
160
+ {%- if role != 'tool' -%}
161
+ {%- if ns.prev_message_type != 'tool_response' -%}
162
+ {{- '<start_of_turn>' + role + '\n' }}
163
+ {%- endif -%}
164
+ {%- set ns.prev_message_type = None -%}
165
+ {%- if 'content' in message and message['content'] is not none -%}
166
+ {%- if message['content'] is string -%}
167
+ {{ message['content'] | trim }}
168
+ {%- elif message['content'] is sequence -%}
169
+ {%- for item in message['content'] -%}
170
+ {%- if item['type'] == 'image' -%}
171
+ {{ '<start_of_image>' }}
172
+ {%- elif item['type'] == 'text' -%}
173
+ {{ item['text'] | trim }}
174
+ {%- endif -%}
175
+ {%- endfor -%}
176
+ {%- else -%}
177
+ {{ raise_exception("Invalid content type in user/assistant message") }}
178
+ {%- endif -%}
179
+ {%- set ns.prev_message_type = 'content' -%}
180
+ {%- endif -%}
181
+ {%- if 'tool_calls' in message and message['tool_calls'] and message['tool_calls'] is iterable -%}
182
+ {#- Tool Calls -#}
183
+ {%- for tool_call in message['tool_calls'] -%}
184
+ {% set function = tool_call['function'] %}
185
+ {{- '<start_function_call>call:' + function['name'] + '{' -}}
186
+ {%- if 'arguments' in function -%}
187
+ {%- if function['arguments'] is mapping -%}
188
+ {%- set ns = namespace(found_first=false) -%}
189
+ {%- for key, value in function['arguments'] | dictsort -%}
190
+ {%- if ns.found_first %},{% endif -%}
191
+ {%- set ns.found_first = true -%}
192
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
193
+ {%- endfor -%}
194
+ {%- elif function['arguments'] is string -%}
195
+ {# This handles string-JSON, just in case #}
196
+ {{ function['arguments'] }}
197
+ {%- endif %}
198
+ {%- endif -%}
199
+ {{- '}<end_function_call>' -}}
200
+ {%- endfor -%}
201
+ {%- if loop.last -%}
202
+ {{ '<start_function_response>' }}
203
+ {%- endif -%}
204
+ {%- set ns.prev_message_type = 'tool_call' -%}
205
+ {%- endif -%}
206
+ {%- else -%}
207
+ {#- Tool Responses -#}
208
+ {%- if 'content' in message and message['content'] -%}
209
+ {%- if message['content'] is mapping -%}
210
+ {%- if 'name' in message['content'] and 'response' in message['content'] -%}
211
+ {{ '<start_function_response>response:' + message['content']['name'] | trim + '{' }}
212
+ {%- set response_ns = namespace(found_first=false) -%}
213
+ {%- for key, value in message['content']['response'] | dictsort -%}
214
+ {%- if response_ns.found_first %},{% endif -%}
215
+ {%- set response_ns.found_first = true -%}
216
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
217
+ {%- endfor -%}
218
+ {{- '}<end_function_response>' -}}
219
+ {%- elif 'name' in message -%}
220
+ {{ '<start_function_response>response:' + message['name'] | trim + '{' }}
221
+ {%- set response_ns = namespace(found_first=false) -%}
222
+ {%- for key, value in message['content'] | dictsort -%}
223
+ {%- if response_ns.found_first %},{% endif -%}
224
+ {%- set response_ns.found_first = true -%}
225
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
226
+ {%- endfor -%}
227
+ {{- '}<end_function_response>' -}}
228
+ {%- else -%}
229
+ {{ raise_exception("Invalid tool response mapping: must contain 'name' and 'response' keys, or 'name' must be in the message.") }}
230
+ {%- endif -%}
231
+ {%- elif message['content'] is string -%}
232
+ {%- if 'name' in message -%}
233
+ {{ '<start_function_response>response:' + message['name'] | trim + '{value:' + format_argument(message['content'], escape_keys=False) + '}<end_function_response>' }}
234
+ {%- else -%}
235
+ {{ raise_exception("Invalid tool response: 'name' must be provided.") }}
236
+ {%- endif -%}
237
+ {%- elif message['content'] is sequence -%}
238
+ {%- for item in message['content'] -%}
239
+ {%- if item is mapping -%}
240
+ {%- if 'name' in item and 'response' in item -%}
241
+ {{ '<start_function_response>response:' + item['name'] | trim + '{' }}
242
+ {%- set response_ns = namespace(found_first=false) -%}
243
+ {%- for key, value in item['response'] | dictsort -%}
244
+ {%- if response_ns.found_first %},{% endif -%}
245
+ {%- set response_ns.found_first = true -%}
246
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
247
+ {%- endfor -%}
248
+ {{- '}<end_function_response>' -}}
249
+ {%- elif 'name' in message -%}
250
+ {{ '<start_function_response>response:' + message['name'] | trim + '{' }}
251
+ {%- set response_ns = namespace(found_first=false) -%}
252
+ {%- for key, value in item | dictsort -%}
253
+ {%- if response_ns.found_first %},{% endif -%}
254
+ {%- set response_ns.found_first = true -%}
255
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
256
+ {%- endfor -%}
257
+ {{- '}<end_function_response>' -}}
258
+ {%- else -%}
259
+ {{ raise_exception("Invalid tool response mapping: must contain 'name' and 'response' keys, or 'name' must be in the message.") }}
260
+ {%- endif -%}
261
+ {%- else -%}
262
+ {{ raise_exception("Invalid tool response message: multiple responses must all be mappings") }}
263
+ {%- endif -%}
264
+ {%- endfor -%}
265
+ {%- else -%}
266
+ {{ raise_exception("Invalid content type in tool message: must be mapping, sequence of mappings, or string.") }}
267
+ {%- endif -%}
268
+ {%- endif -%}
269
+ {%- set ns.prev_message_type = 'tool_response' -%}
270
+ {%- endif -%}
271
+ {%- if ns.prev_message_type not in ['tool_call', 'tool_response'] -%}
272
+ {{ '<end_of_turn>\n' }}
273
+ {%- endif -%}
274
+ {%- endfor -%}
275
+ {%- if add_generation_prompt -%}
276
+ {%- if ns.prev_message_type != 'tool_response' -%}
277
+ {{- '<start_of_turn>model\n' -}}
278
+ {%- endif -%}
279
+ {%- endif -%}
checkpoint-50/config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_sliding_window_pattern": 6,
3
+ "architectures": [
4
+ "Gemma3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_logit_softcapping": null,
9
+ "bos_token_id": 2,
10
+ "dtype": "float32",
11
+ "eos_token_id": 106,
12
+ "final_logit_softcapping": null,
13
+ "head_dim": 256,
14
+ "hidden_activation": "gelu_pytorch_tanh",
15
+ "hidden_size": 640,
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 2048,
18
+ "layer_types": [
19
+ "sliding_attention",
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "full_attention",
25
+ "sliding_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "full_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention"
37
+ ],
38
+ "max_position_embeddings": 32768,
39
+ "model_type": "gemma3_text",
40
+ "num_attention_heads": 4,
41
+ "num_hidden_layers": 18,
42
+ "num_key_value_heads": 1,
43
+ "pad_token_id": 0,
44
+ "query_pre_attn_scalar": 256,
45
+ "rms_norm_eps": 1e-06,
46
+ "rope_local_base_freq": 10000.0,
47
+ "rope_scaling": null,
48
+ "rope_theta": 1000000.0,
49
+ "sliding_window": 512,
50
+ "transformers_version": "4.57.1",
51
+ "use_bidirectional_attention": false,
52
+ "use_cache": false,
53
+ "vocab_size": 262144
54
+ }
checkpoint-50/generation_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 106,
7
+ 1,
8
+ 50,
9
+ 106
10
+ ],
11
+ "pad_token_id": 0,
12
+ "top_k": 64,
13
+ "top_p": 0.95,
14
+ "transformers_version": "4.57.1"
15
+ }
checkpoint-50/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70c84098f6184e323e0ef3f17bb3acadf76f9b9e6594ff0f1c61f5e181cabbbb
3
+ size 1072419256
checkpoint-50/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dcd1b980b145de2b9005a186315d27f006ee778952f74df2b636ab60f7c5e4e7
3
+ size 2144987083
checkpoint-50/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4a9f217e852f439efa6bd32fde98d6867f11aa6ea13ddc021ba10af6a0b0934
3
+ size 14645
checkpoint-50/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e87f0f920b838e6ba8d343e7f107e49478949407e26c58d3cf5cc606fd47c14a
3
+ size 1465
checkpoint-50/special_tokens_map.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "boi_token": "<start_of_image>",
3
+ "bos_token": {
4
+ "content": "<bos>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "eoi_token": "<end_of_image>",
11
+ "eos_token": {
12
+ "content": "<end_of_turn>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "image_token": "<image_soft_token>",
19
+ "pad_token": {
20
+ "content": "<pad>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false
25
+ },
26
+ "sfr_token": "<start_function_response>",
27
+ "unk_token": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false
33
+ }
34
+ }
checkpoint-50/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6b09a0b4a803ad453063ca4bb49a784540e8120004e2450e025df2b27d41fb2
3
+ size 33384899
checkpoint-50/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa009fcbc3589a9904d30d04834094fea4653c2ac6d2de2cd1262d4f7a50ceb3
3
+ size 4689144
checkpoint-50/tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-50/trainer_state.json ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 2.280701754385965,
6
+ "eval_steps": 500,
7
+ "global_step": 50,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.23391812865497075,
14
+ "grad_norm": 19.255443572998047,
15
+ "learning_rate": 2.857142857142857e-05,
16
+ "loss": 0.8241,
17
+ "step": 5
18
+ },
19
+ {
20
+ "epoch": 0.4678362573099415,
21
+ "grad_norm": 15.90062141418457,
22
+ "learning_rate": 4.985837000525343e-05,
23
+ "loss": 0.1671,
24
+ "step": 10
25
+ },
26
+ {
27
+ "epoch": 0.7017543859649122,
28
+ "grad_norm": 5.5942206382751465,
29
+ "learning_rate": 4.8283404436308464e-05,
30
+ "loss": 0.085,
31
+ "step": 15
32
+ },
33
+ {
34
+ "epoch": 0.935672514619883,
35
+ "grad_norm": 10.160480499267578,
36
+ "learning_rate": 4.5067790948274094e-05,
37
+ "loss": 0.0862,
38
+ "step": 20
39
+ },
40
+ {
41
+ "epoch": 1.1403508771929824,
42
+ "grad_norm": 0.6265833973884583,
43
+ "learning_rate": 4.04381153736548e-05,
44
+ "loss": 0.0485,
45
+ "step": 25
46
+ },
47
+ {
48
+ "epoch": 1.3742690058479532,
49
+ "grad_norm": 0.9624802470207214,
50
+ "learning_rate": 3.472060438683302e-05,
51
+ "loss": 0.0484,
52
+ "step": 30
53
+ },
54
+ {
55
+ "epoch": 1.608187134502924,
56
+ "grad_norm": 0.569491982460022,
57
+ "learning_rate": 2.8318138182093052e-05,
58
+ "loss": 0.0468,
59
+ "step": 35
60
+ },
61
+ {
62
+ "epoch": 1.8421052631578947,
63
+ "grad_norm": 0.3710443377494812,
64
+ "learning_rate": 2.1681861817906954e-05,
65
+ "loss": 0.0458,
66
+ "step": 40
67
+ },
68
+ {
69
+ "epoch": 2.046783625730994,
70
+ "grad_norm": 0.1184922382235527,
71
+ "learning_rate": 1.5279395613166986e-05,
72
+ "loss": 0.0455,
73
+ "step": 45
74
+ },
75
+ {
76
+ "epoch": 2.280701754385965,
77
+ "grad_norm": 0.21576550602912903,
78
+ "learning_rate": 9.561884626345205e-06,
79
+ "loss": 0.0439,
80
+ "step": 50
81
+ }
82
+ ],
83
+ "logging_steps": 5,
84
+ "max_steps": 66,
85
+ "num_input_tokens_seen": 0,
86
+ "num_train_epochs": 3,
87
+ "save_steps": 50,
88
+ "stateful_callbacks": {
89
+ "TrainerControl": {
90
+ "args": {
91
+ "should_epoch_stop": false,
92
+ "should_evaluate": false,
93
+ "should_log": false,
94
+ "should_save": true,
95
+ "should_training_stop": false
96
+ },
97
+ "attributes": {}
98
+ }
99
+ },
100
+ "total_flos": 312535605043200.0,
101
+ "train_batch_size": 2,
102
+ "trial_name": null,
103
+ "trial_params": null
104
+ }
checkpoint-50/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f885fe321b69c9af13ef56d2f6442f1c14f44ed068efcbb9522d2fab6e7f40b
3
+ size 6353
checkpoint-66/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<end_of_image>": 262145,
3
+ "<image_soft_token>": 262144
4
+ }
checkpoint-66/chat_template.jinja ADDED
@@ -0,0 +1,279 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- macro format_parameters(properties, required) -%}
2
+ {%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%}
3
+ {%- set ns = namespace(found_first=false) -%}
4
+ {%- for key, value in properties | dictsort -%}
5
+ {%- if key not in standard_keys -%}
6
+ {%- if ns.found_first %},{% endif -%}
7
+ {%- set ns.found_first = true -%}
8
+ {{- key }}:{description:<escape>{{ value['description'] }}<escape>
9
+ {%- if value['type'] | upper == 'STRING' -%}
10
+ {%- if value['enum'] -%}
11
+ ,enum:{{ format_argument(value['enum']) }}
12
+ {%- endif -%}
13
+ {%- elif value['type'] | upper == 'OBJECT' -%}
14
+ ,properties:{
15
+ {%- if value['properties'] is defined and value['properties'] is mapping -%}
16
+ {{- format_parameters(value['properties'], value['required'] | default([])) -}}
17
+ {%- elif value is mapping -%}
18
+ {{- format_parameters(value, value['required'] | default([])) -}}
19
+ {%- endif -%}
20
+ }
21
+ {%- if value['required'] -%}
22
+ ,required:[
23
+ {%- for item in value['required'] | default([]) -%}
24
+ <escape>{{- item -}}<escape>
25
+ {%- if not loop.last %},{% endif -%}
26
+ {%- endfor -%}
27
+ ]
28
+ {%- endif -%}
29
+ {%- elif value['type'] | upper == 'ARRAY' -%}
30
+ {%- if value['items'] is mapping and value['items'] -%}
31
+ ,items:{
32
+ {%- set ns_items = namespace(found_first=false) -%}
33
+ {%- for item_key, item_value in value['items'] | dictsort -%}
34
+ {%- if item_value is not none -%}
35
+ {%- if ns_items.found_first %},{% endif -%}
36
+ {%- set ns_items.found_first = true -%}
37
+ {%- if item_key == 'properties' -%}
38
+ properties:{
39
+ {%- if item_value is mapping -%}
40
+ {{- format_parameters(item_value, value['items']['required'] | default([])) -}}
41
+ {%- endif -%}
42
+ }
43
+ {%- elif item_key == 'required' -%}
44
+ required:[
45
+ {%- for req_item in item_value -%}
46
+ <escape>{{- req_item -}}<escape>
47
+ {%- if not loop.last %},{% endif -%}
48
+ {%- endfor -%}
49
+ ]
50
+ {%- elif item_key == 'type' -%}
51
+ {%- if item_value is string -%}
52
+ type:{{ format_argument(item_value | upper) }}
53
+ {%- else -%}
54
+ type:{{ format_argument(item_value | map('upper') | list) }}
55
+ {%- endif -%}
56
+ {%- else -%}
57
+ {{ item_key }}:{{ format_argument(item_value) }}
58
+ {%- endif -%}
59
+ {%- endif -%}
60
+ {%- endfor -%}
61
+ }
62
+ {%- endif -%}
63
+ {%- endif -%}
64
+ ,type:<escape>{{ value['type'] | upper }}<escape>}
65
+ {%- endif -%}
66
+ {%- endfor -%}
67
+ {%- endmacro -%}
68
+ {% macro format_function_declaration(tool_data) -%}
69
+ declaration:{{- tool_data['function']['name'] -}}
70
+ {description:<escape>{{- tool_data['function']['description'] -}}<escape>
71
+ {%- set params = tool_data['function']['parameters'] -%}
72
+ {%- if params -%}
73
+ ,parameters:{
74
+ {%- if params['properties'] -%}
75
+ properties:{ {{- format_parameters(params['properties'], params['required']) -}} },
76
+ {%- endif -%}
77
+ {%- if params['required'] -%}
78
+ required:[
79
+ {%- for item in params['required'] -%}
80
+ <escape>{{- item -}}<escape>
81
+ {{- ',' if not loop.last -}}
82
+ {%- endfor -%}
83
+ ],
84
+ {%- endif -%}
85
+ {%- if params['type'] -%}
86
+ type:<escape>{{- params['type'] | upper -}}<escape>}
87
+ {%- endif -%}
88
+ {%- endif -%}
89
+ }
90
+ {%- endmacro -%}
91
+ {% macro format_argument(argument, escape_keys=True) -%}
92
+ {%- if argument is string -%}
93
+ {{- '<escape>' + argument + '<escape>' -}}
94
+ {%- elif argument is boolean -%}
95
+ {%- if argument -%}
96
+ {{- 'true' -}}
97
+ {%- else -%}
98
+ {{- 'false' -}}
99
+ {%- endif -%}
100
+ {%- elif argument is mapping -%}
101
+ {{- '{' -}}
102
+ {%- set ns = namespace(found_first=false) -%}
103
+ {%- for key, value in argument | dictsort -%}
104
+ {%- if ns.found_first %},{% endif -%}
105
+ {%- set ns.found_first = true -%}
106
+ {%- if escape_keys -%}
107
+ {{- '<escape>' + key + '<escape>' -}}
108
+ {%- else -%}
109
+ {{- key -}}
110
+ {%- endif -%}
111
+ :{{- format_argument(value, escape_keys=escape_keys) -}}
112
+ {%- endfor -%}
113
+ {{- '}' -}}
114
+ {%- elif argument is sequence -%}
115
+ {{- '[' -}}
116
+ {%- for item in argument -%}
117
+ {{- format_argument(item, escape_keys=escape_keys) -}}
118
+ {%- if not loop.last %},{% endif -%}
119
+ {%- endfor -%}
120
+ {{- ']' -}}
121
+ {%- else -%}
122
+ {{- argument -}}
123
+ {%- endif -%}
124
+ {%- endmacro -%}
125
+ {{ bos_token }}
126
+ {%- set ns = namespace(prev_message_type=None) -%}
127
+ {#- Tool Declarations -#}
128
+ {%- set loop_messages = messages -%}
129
+ {%- if tools or messages[0]['role'] == 'system' or messages[0]['role'] == 'developer' -%}
130
+ {{- '<start_of_turn>developer\n' -}}
131
+ {%- if messages[0]['role'] == 'system' or messages[0]['role'] == 'developer' -%}
132
+ {%- if messages[0]['content'] is string -%}
133
+ {{- messages[0]['content'] | trim -}}
134
+ {%- elif messages[0]['content'] is sequence -%}
135
+ {%- for item in messages[0]['content'] -%}
136
+ {%- if item['type'] == 'text' -%}
137
+ {{- item['text'] | trim -}}
138
+ {%- endif -%}
139
+ {%- endfor -%}
140
+ {%- endif -%}
141
+ {%- set loop_messages = messages[1:] -%}
142
+ {%- endif -%}
143
+ {%- if tools -%}
144
+ {%- for tool in tools %}
145
+ {{- '<start_function_declaration>' -}}
146
+ {{- format_function_declaration(tool) | trim }}
147
+ {{- '<end_function_declaration>' -}}
148
+ {%- endfor %}
149
+ {%- endif -%}
150
+ {{- '<end_of_turn>\n' }}
151
+ {%- endif %}
152
+ {#- Loop through messages. -#}
153
+ {%- for message in loop_messages -%}
154
+ {%- if (message['role'] == 'assistant') -%}
155
+ {#- Rename "assistant" to "model". -#}
156
+ {%- set role = "model" -%}
157
+ {%- else -%}
158
+ {%- set role = message['role'] -%}
159
+ {%- endif -%}
160
+ {%- if role != 'tool' -%}
161
+ {%- if ns.prev_message_type != 'tool_response' -%}
162
+ {{- '<start_of_turn>' + role + '\n' }}
163
+ {%- endif -%}
164
+ {%- set ns.prev_message_type = None -%}
165
+ {%- if 'content' in message and message['content'] is not none -%}
166
+ {%- if message['content'] is string -%}
167
+ {{ message['content'] | trim }}
168
+ {%- elif message['content'] is sequence -%}
169
+ {%- for item in message['content'] -%}
170
+ {%- if item['type'] == 'image' -%}
171
+ {{ '<start_of_image>' }}
172
+ {%- elif item['type'] == 'text' -%}
173
+ {{ item['text'] | trim }}
174
+ {%- endif -%}
175
+ {%- endfor -%}
176
+ {%- else -%}
177
+ {{ raise_exception("Invalid content type in user/assistant message") }}
178
+ {%- endif -%}
179
+ {%- set ns.prev_message_type = 'content' -%}
180
+ {%- endif -%}
181
+ {%- if 'tool_calls' in message and message['tool_calls'] and message['tool_calls'] is iterable -%}
182
+ {#- Tool Calls -#}
183
+ {%- for tool_call in message['tool_calls'] -%}
184
+ {% set function = tool_call['function'] %}
185
+ {{- '<start_function_call>call:' + function['name'] + '{' -}}
186
+ {%- if 'arguments' in function -%}
187
+ {%- if function['arguments'] is mapping -%}
188
+ {%- set ns = namespace(found_first=false) -%}
189
+ {%- for key, value in function['arguments'] | dictsort -%}
190
+ {%- if ns.found_first %},{% endif -%}
191
+ {%- set ns.found_first = true -%}
192
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
193
+ {%- endfor -%}
194
+ {%- elif function['arguments'] is string -%}
195
+ {# This handles string-JSON, just in case #}
196
+ {{ function['arguments'] }}
197
+ {%- endif %}
198
+ {%- endif -%}
199
+ {{- '}<end_function_call>' -}}
200
+ {%- endfor -%}
201
+ {%- if loop.last -%}
202
+ {{ '<start_function_response>' }}
203
+ {%- endif -%}
204
+ {%- set ns.prev_message_type = 'tool_call' -%}
205
+ {%- endif -%}
206
+ {%- else -%}
207
+ {#- Tool Responses -#}
208
+ {%- if 'content' in message and message['content'] -%}
209
+ {%- if message['content'] is mapping -%}
210
+ {%- if 'name' in message['content'] and 'response' in message['content'] -%}
211
+ {{ '<start_function_response>response:' + message['content']['name'] | trim + '{' }}
212
+ {%- set response_ns = namespace(found_first=false) -%}
213
+ {%- for key, value in message['content']['response'] | dictsort -%}
214
+ {%- if response_ns.found_first %},{% endif -%}
215
+ {%- set response_ns.found_first = true -%}
216
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
217
+ {%- endfor -%}
218
+ {{- '}<end_function_response>' -}}
219
+ {%- elif 'name' in message -%}
220
+ {{ '<start_function_response>response:' + message['name'] | trim + '{' }}
221
+ {%- set response_ns = namespace(found_first=false) -%}
222
+ {%- for key, value in message['content'] | dictsort -%}
223
+ {%- if response_ns.found_first %},{% endif -%}
224
+ {%- set response_ns.found_first = true -%}
225
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
226
+ {%- endfor -%}
227
+ {{- '}<end_function_response>' -}}
228
+ {%- else -%}
229
+ {{ raise_exception("Invalid tool response mapping: must contain 'name' and 'response' keys, or 'name' must be in the message.") }}
230
+ {%- endif -%}
231
+ {%- elif message['content'] is string -%}
232
+ {%- if 'name' in message -%}
233
+ {{ '<start_function_response>response:' + message['name'] | trim + '{value:' + format_argument(message['content'], escape_keys=False) + '}<end_function_response>' }}
234
+ {%- else -%}
235
+ {{ raise_exception("Invalid tool response: 'name' must be provided.") }}
236
+ {%- endif -%}
237
+ {%- elif message['content'] is sequence -%}
238
+ {%- for item in message['content'] -%}
239
+ {%- if item is mapping -%}
240
+ {%- if 'name' in item and 'response' in item -%}
241
+ {{ '<start_function_response>response:' + item['name'] | trim + '{' }}
242
+ {%- set response_ns = namespace(found_first=false) -%}
243
+ {%- for key, value in item['response'] | dictsort -%}
244
+ {%- if response_ns.found_first %},{% endif -%}
245
+ {%- set response_ns.found_first = true -%}
246
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
247
+ {%- endfor -%}
248
+ {{- '}<end_function_response>' -}}
249
+ {%- elif 'name' in message -%}
250
+ {{ '<start_function_response>response:' + message['name'] | trim + '{' }}
251
+ {%- set response_ns = namespace(found_first=false) -%}
252
+ {%- for key, value in item | dictsort -%}
253
+ {%- if response_ns.found_first %},{% endif -%}
254
+ {%- set response_ns.found_first = true -%}
255
+ {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
256
+ {%- endfor -%}
257
+ {{- '}<end_function_response>' -}}
258
+ {%- else -%}
259
+ {{ raise_exception("Invalid tool response mapping: must contain 'name' and 'response' keys, or 'name' must be in the message.") }}
260
+ {%- endif -%}
261
+ {%- else -%}
262
+ {{ raise_exception("Invalid tool response message: multiple responses must all be mappings") }}
263
+ {%- endif -%}
264
+ {%- endfor -%}
265
+ {%- else -%}
266
+ {{ raise_exception("Invalid content type in tool message: must be mapping, sequence of mappings, or string.") }}
267
+ {%- endif -%}
268
+ {%- endif -%}
269
+ {%- set ns.prev_message_type = 'tool_response' -%}
270
+ {%- endif -%}
271
+ {%- if ns.prev_message_type not in ['tool_call', 'tool_response'] -%}
272
+ {{ '<end_of_turn>\n' }}
273
+ {%- endif -%}
274
+ {%- endfor -%}
275
+ {%- if add_generation_prompt -%}
276
+ {%- if ns.prev_message_type != 'tool_response' -%}
277
+ {{- '<start_of_turn>model\n' -}}
278
+ {%- endif -%}
279
+ {%- endif -%}
checkpoint-66/config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_sliding_window_pattern": 6,
3
+ "architectures": [
4
+ "Gemma3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_logit_softcapping": null,
9
+ "bos_token_id": 2,
10
+ "dtype": "float32",
11
+ "eos_token_id": 106,
12
+ "final_logit_softcapping": null,
13
+ "head_dim": 256,
14
+ "hidden_activation": "gelu_pytorch_tanh",
15
+ "hidden_size": 640,
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 2048,
18
+ "layer_types": [
19
+ "sliding_attention",
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "full_attention",
25
+ "sliding_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "full_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention"
37
+ ],
38
+ "max_position_embeddings": 32768,
39
+ "model_type": "gemma3_text",
40
+ "num_attention_heads": 4,
41
+ "num_hidden_layers": 18,
42
+ "num_key_value_heads": 1,
43
+ "pad_token_id": 0,
44
+ "query_pre_attn_scalar": 256,
45
+ "rms_norm_eps": 1e-06,
46
+ "rope_local_base_freq": 10000.0,
47
+ "rope_scaling": null,
48
+ "rope_theta": 1000000.0,
49
+ "sliding_window": 512,
50
+ "transformers_version": "4.57.1",
51
+ "use_bidirectional_attention": false,
52
+ "use_cache": false,
53
+ "vocab_size": 262144
54
+ }
checkpoint-66/generation_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 106,
7
+ 1,
8
+ 50,
9
+ 106
10
+ ],
11
+ "pad_token_id": 0,
12
+ "top_k": 64,
13
+ "top_p": 0.95,
14
+ "transformers_version": "4.57.1"
15
+ }
checkpoint-66/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63d68c3f7791a1e6f9493851e11769c39ba8152c3581916205101b60850709c5
3
+ size 1072419256
checkpoint-66/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b20b4b36342ed9574e02d0ad1fce0e370054a8bea475c8b44231cf452004bf1
3
+ size 2144987083
checkpoint-66/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4a9f217e852f439efa6bd32fde98d6867f11aa6ea13ddc021ba10af6a0b0934
3
+ size 14645
checkpoint-66/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e53313a4c270fc58d84709a54609c670b94786a66a6aae599442e6218bd5fc22
3
+ size 1465
checkpoint-66/special_tokens_map.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "boi_token": "<start_of_image>",
3
+ "bos_token": {
4
+ "content": "<bos>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "eoi_token": "<end_of_image>",
11
+ "eos_token": {
12
+ "content": "<end_of_turn>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "image_token": "<image_soft_token>",
19
+ "pad_token": {
20
+ "content": "<pad>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false
25
+ },
26
+ "sfr_token": "<start_function_response>",
27
+ "unk_token": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false
33
+ }
34
+ }
checkpoint-66/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6b09a0b4a803ad453063ca4bb49a784540e8120004e2450e025df2b27d41fb2
3
+ size 33384899
checkpoint-66/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa009fcbc3589a9904d30d04834094fea4653c2ac6d2de2cd1262d4f7a50ceb3
3
+ size 4689144
checkpoint-66/tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-66/trainer_state.json ADDED
@@ -0,0 +1,125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 66,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.23391812865497075,
14
+ "grad_norm": 19.255443572998047,
15
+ "learning_rate": 2.857142857142857e-05,
16
+ "loss": 0.8241,
17
+ "step": 5
18
+ },
19
+ {
20
+ "epoch": 0.4678362573099415,
21
+ "grad_norm": 15.90062141418457,
22
+ "learning_rate": 4.985837000525343e-05,
23
+ "loss": 0.1671,
24
+ "step": 10
25
+ },
26
+ {
27
+ "epoch": 0.7017543859649122,
28
+ "grad_norm": 5.5942206382751465,
29
+ "learning_rate": 4.8283404436308464e-05,
30
+ "loss": 0.085,
31
+ "step": 15
32
+ },
33
+ {
34
+ "epoch": 0.935672514619883,
35
+ "grad_norm": 10.160480499267578,
36
+ "learning_rate": 4.5067790948274094e-05,
37
+ "loss": 0.0862,
38
+ "step": 20
39
+ },
40
+ {
41
+ "epoch": 1.1403508771929824,
42
+ "grad_norm": 0.6265833973884583,
43
+ "learning_rate": 4.04381153736548e-05,
44
+ "loss": 0.0485,
45
+ "step": 25
46
+ },
47
+ {
48
+ "epoch": 1.3742690058479532,
49
+ "grad_norm": 0.9624802470207214,
50
+ "learning_rate": 3.472060438683302e-05,
51
+ "loss": 0.0484,
52
+ "step": 30
53
+ },
54
+ {
55
+ "epoch": 1.608187134502924,
56
+ "grad_norm": 0.569491982460022,
57
+ "learning_rate": 2.8318138182093052e-05,
58
+ "loss": 0.0468,
59
+ "step": 35
60
+ },
61
+ {
62
+ "epoch": 1.8421052631578947,
63
+ "grad_norm": 0.3710443377494812,
64
+ "learning_rate": 2.1681861817906954e-05,
65
+ "loss": 0.0458,
66
+ "step": 40
67
+ },
68
+ {
69
+ "epoch": 2.046783625730994,
70
+ "grad_norm": 0.1184922382235527,
71
+ "learning_rate": 1.5279395613166986e-05,
72
+ "loss": 0.0455,
73
+ "step": 45
74
+ },
75
+ {
76
+ "epoch": 2.280701754385965,
77
+ "grad_norm": 0.21576550602912903,
78
+ "learning_rate": 9.561884626345205e-06,
79
+ "loss": 0.0439,
80
+ "step": 50
81
+ },
82
+ {
83
+ "epoch": 2.5146198830409356,
84
+ "grad_norm": 0.17528507113456726,
85
+ "learning_rate": 4.932209051725914e-06,
86
+ "loss": 0.0442,
87
+ "step": 55
88
+ },
89
+ {
90
+ "epoch": 2.7485380116959064,
91
+ "grad_norm": 0.1674930304288864,
92
+ "learning_rate": 1.7165955636915392e-06,
93
+ "loss": 0.0438,
94
+ "step": 60
95
+ },
96
+ {
97
+ "epoch": 2.982456140350877,
98
+ "grad_norm": 0.1652153730392456,
99
+ "learning_rate": 1.4162999474657268e-07,
100
+ "loss": 0.0451,
101
+ "step": 65
102
+ }
103
+ ],
104
+ "logging_steps": 5,
105
+ "max_steps": 66,
106
+ "num_input_tokens_seen": 0,
107
+ "num_train_epochs": 3,
108
+ "save_steps": 50,
109
+ "stateful_callbacks": {
110
+ "TrainerControl": {
111
+ "args": {
112
+ "should_epoch_stop": false,
113
+ "should_evaluate": false,
114
+ "should_log": false,
115
+ "should_save": true,
116
+ "should_training_stop": true
117
+ },
118
+ "attributes": {}
119
+ }
120
+ },
121
+ "total_flos": 411612762796032.0,
122
+ "train_batch_size": 2,
123
+ "trial_name": null,
124
+ "trial_params": null
125
+ }
checkpoint-66/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f885fe321b69c9af13ef56d2f6442f1c14f44ed068efcbb9522d2fab6e7f40b
3
+ size 6353
config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_sliding_window_pattern": 6,
3
+ "architectures": [
4
+ "Gemma3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_logit_softcapping": null,
9
+ "bos_token_id": 2,
10
+ "dtype": "float32",
11
+ "eos_token_id": 106,
12
+ "final_logit_softcapping": null,
13
+ "head_dim": 256,
14
+ "hidden_activation": "gelu_pytorch_tanh",
15
+ "hidden_size": 640,
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 2048,
18
+ "layer_types": [
19
+ "sliding_attention",
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "full_attention",
25
+ "sliding_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "full_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention"
37
+ ],
38
+ "max_position_embeddings": 32768,
39
+ "model_type": "gemma3_text",
40
+ "num_attention_heads": 4,
41
+ "num_hidden_layers": 18,
42
+ "num_key_value_heads": 1,
43
+ "pad_token_id": 0,
44
+ "query_pre_attn_scalar": 256,
45
+ "rms_norm_eps": 1e-06,
46
+ "rope_local_base_freq": 10000.0,
47
+ "rope_scaling": null,
48
+ "rope_theta": 1000000.0,
49
+ "sliding_window": 512,
50
+ "transformers_version": "4.57.1",
51
+ "use_bidirectional_attention": false,
52
+ "use_cache": false,
53
+ "vocab_size": 262144
54
+ }
generation_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 106,
7
+ 1,
8
+ 50,
9
+ 106
10
+ ],
11
+ "pad_token_id": 0,
12
+ "top_k": 64,
13
+ "top_p": 0.95,
14
+ "transformers_version": "4.57.1"
15
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63d68c3f7791a1e6f9493851e11769c39ba8152c3581916205101b60850709c5
3
+ size 1072419256
runs/Jan02_13-47-02_074ad1548a3d/events.out.tfevents.1767361678.074ad1548a3d.3005.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9ed9d4d3e448aa08974a3cc8c3d52fa8a96cf91487679244d73b11cca3a41615
3
+ size 6268
runs/Jan02_13-50-09_074ad1548a3d/events.out.tfevents.1767361816.074ad1548a3d.4055.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78b27d14a45080677beae604f2f2dfd3ef1913fa781ffb20b64299f16fd32078
3
+ size 6265
runs/Jan02_13-53-13_074ad1548a3d/events.out.tfevents.1767362039.074ad1548a3d.4947.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3dae6745e429f566515e437d9116382762adb789549aa24ebb125cf7468e5d26
3
+ size 9304
special_tokens_map.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "boi_token": "<start_of_image>",
3
+ "bos_token": {
4
+ "content": "<bos>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "eoi_token": "<end_of_image>",
11
+ "eos_token": {
12
+ "content": "<end_of_turn>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "image_token": "<image_soft_token>",
19
+ "pad_token": {
20
+ "content": "<pad>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false
25
+ },
26
+ "sfr_token": "<start_function_response>",
27
+ "unk_token": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false
33
+ }
34
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6b09a0b4a803ad453063ca4bb49a784540e8120004e2450e025df2b27d41fb2
3
+ size 33384899
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa009fcbc3589a9904d30d04834094fea4653c2ac6d2de2cd1262d4f7a50ceb3
3
+ size 4689144
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 411612762796032.0,
4
+ "train_loss": 0.12000229429792274,
5
+ "train_runtime": 714.9914,
6
+ "train_samples_per_second": 1.435,
7
+ "train_steps_per_second": 0.092
8
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 5, "total_steps": 66, "loss": 0.8241, "lr": 2.857142857142857e-05, "epoch": 0.23391812865497075, "percentage": 7.58, "elapsed_time": "0:00:51", "remaining_time": "0:10:31"}
2
+ {"current_steps": 10, "total_steps": 66, "loss": 0.1671, "lr": 4.985837000525343e-05, "epoch": 0.4678362573099415, "percentage": 15.15, "elapsed_time": "0:01:43", "remaining_time": "0:09:39"}
3
+ {"current_steps": 15, "total_steps": 66, "loss": 0.085, "lr": 4.8283404436308464e-05, "epoch": 0.7017543859649122, "percentage": 22.73, "elapsed_time": "0:02:34", "remaining_time": "0:08:44"}
4
+ {"current_steps": 20, "total_steps": 66, "loss": 0.0862, "lr": 4.5067790948274094e-05, "epoch": 0.935672514619883, "percentage": 30.3, "elapsed_time": "0:03:25", "remaining_time": "0:07:53"}
5
+ {"current_steps": 25, "total_steps": 66, "loss": 0.0485, "lr": 4.04381153736548e-05, "epoch": 1.1403508771929824, "percentage": 37.88, "elapsed_time": "0:04:10", "remaining_time": "0:06:50"}
6
+ {"current_steps": 30, "total_steps": 66, "loss": 0.0484, "lr": 3.472060438683302e-05, "epoch": 1.3742690058479532, "percentage": 45.45, "elapsed_time": "0:05:01", "remaining_time": "0:06:01"}
7
+ {"current_steps": 35, "total_steps": 66, "loss": 0.0468, "lr": 2.8318138182093052e-05, "epoch": 1.608187134502924, "percentage": 53.03, "elapsed_time": "0:05:52", "remaining_time": "0:05:12"}
8
+ {"current_steps": 40, "total_steps": 66, "loss": 0.0458, "lr": 2.1681861817906954e-05, "epoch": 1.8421052631578947, "percentage": 60.61, "elapsed_time": "0:06:44", "remaining_time": "0:04:22"}
9
+ {"current_steps": 45, "total_steps": 66, "loss": 0.0455, "lr": 1.5279395613166986e-05, "epoch": 2.046783625730994, "percentage": 68.18, "elapsed_time": "0:07:29", "remaining_time": "0:03:29"}
10
+ {"current_steps": 50, "total_steps": 66, "loss": 0.0439, "lr": 9.561884626345205e-06, "epoch": 2.280701754385965, "percentage": 75.76, "elapsed_time": "0:08:19", "remaining_time": "0:02:39"}
11
+ {"current_steps": 55, "total_steps": 66, "loss": 0.0442, "lr": 4.932209051725914e-06, "epoch": 2.5146198830409356, "percentage": 83.33, "elapsed_time": "0:09:37", "remaining_time": "0:01:55"}
12
+ {"current_steps": 60, "total_steps": 66, "loss": 0.0438, "lr": 1.7165955636915392e-06, "epoch": 2.7485380116959064, "percentage": 90.91, "elapsed_time": "0:10:28", "remaining_time": "0:01:02"}
13
+ {"current_steps": 65, "total_steps": 66, "loss": 0.0451, "lr": 1.4162999474657268e-07, "epoch": 2.982456140350877, "percentage": 98.48, "elapsed_time": "0:11:20", "remaining_time": "0:00:10"}
14
+ {"current_steps": 66, "total_steps": 66, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "0:11:53", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 66,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.23391812865497075,
14
+ "grad_norm": 19.255443572998047,
15
+ "learning_rate": 2.857142857142857e-05,
16
+ "loss": 0.8241,
17
+ "step": 5
18
+ },
19
+ {
20
+ "epoch": 0.4678362573099415,
21
+ "grad_norm": 15.90062141418457,
22
+ "learning_rate": 4.985837000525343e-05,
23
+ "loss": 0.1671,
24
+ "step": 10
25
+ },
26
+ {
27
+ "epoch": 0.7017543859649122,
28
+ "grad_norm": 5.5942206382751465,
29
+ "learning_rate": 4.8283404436308464e-05,
30
+ "loss": 0.085,
31
+ "step": 15
32
+ },
33
+ {
34
+ "epoch": 0.935672514619883,
35
+ "grad_norm": 10.160480499267578,
36
+ "learning_rate": 4.5067790948274094e-05,
37
+ "loss": 0.0862,
38
+ "step": 20
39
+ },
40
+ {
41
+ "epoch": 1.1403508771929824,
42
+ "grad_norm": 0.6265833973884583,
43
+ "learning_rate": 4.04381153736548e-05,
44
+ "loss": 0.0485,
45
+ "step": 25
46
+ },
47
+ {
48
+ "epoch": 1.3742690058479532,
49
+ "grad_norm": 0.9624802470207214,
50
+ "learning_rate": 3.472060438683302e-05,
51
+ "loss": 0.0484,
52
+ "step": 30
53
+ },
54
+ {
55
+ "epoch": 1.608187134502924,
56
+ "grad_norm": 0.569491982460022,
57
+ "learning_rate": 2.8318138182093052e-05,
58
+ "loss": 0.0468,
59
+ "step": 35
60
+ },
61
+ {
62
+ "epoch": 1.8421052631578947,
63
+ "grad_norm": 0.3710443377494812,
64
+ "learning_rate": 2.1681861817906954e-05,
65
+ "loss": 0.0458,
66
+ "step": 40
67
+ },
68
+ {
69
+ "epoch": 2.046783625730994,
70
+ "grad_norm": 0.1184922382235527,
71
+ "learning_rate": 1.5279395613166986e-05,
72
+ "loss": 0.0455,
73
+ "step": 45
74
+ },
75
+ {
76
+ "epoch": 2.280701754385965,
77
+ "grad_norm": 0.21576550602912903,
78
+ "learning_rate": 9.561884626345205e-06,
79
+ "loss": 0.0439,
80
+ "step": 50
81
+ },
82
+ {
83
+ "epoch": 2.5146198830409356,
84
+ "grad_norm": 0.17528507113456726,
85
+ "learning_rate": 4.932209051725914e-06,
86
+ "loss": 0.0442,
87
+ "step": 55
88
+ },
89
+ {
90
+ "epoch": 2.7485380116959064,
91
+ "grad_norm": 0.1674930304288864,
92
+ "learning_rate": 1.7165955636915392e-06,
93
+ "loss": 0.0438,
94
+ "step": 60
95
+ },
96
+ {
97
+ "epoch": 2.982456140350877,
98
+ "grad_norm": 0.1652153730392456,
99
+ "learning_rate": 1.4162999474657268e-07,
100
+ "loss": 0.0451,
101
+ "step": 65
102
+ },
103
+ {
104
+ "epoch": 3.0,
105
+ "step": 66,
106
+ "total_flos": 411612762796032.0,
107
+ "train_loss": 0.12000229429792274,
108
+ "train_runtime": 714.9914,
109
+ "train_samples_per_second": 1.435,
110
+ "train_steps_per_second": 0.092
111
+ }
112
+ ],
113
+ "logging_steps": 5,
114
+ "max_steps": 66,
115
+ "num_input_tokens_seen": 0,
116
+ "num_train_epochs": 3,
117
+ "save_steps": 50,
118
+ "stateful_callbacks": {
119
+ "TrainerControl": {
120
+ "args": {
121
+ "should_epoch_stop": false,
122
+ "should_evaluate": false,
123
+ "should_log": false,
124
+ "should_save": true,
125
+ "should_training_stop": true
126
+ },
127
+ "attributes": {}
128
+ }
129
+ },
130
+ "total_flos": 411612762796032.0,
131
+ "train_batch_size": 2,
132
+ "trial_name": null,
133
+ "trial_params": null
134
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f885fe321b69c9af13ef56d2f6442f1c14f44ed068efcbb9522d2fab6e7f40b
3
+ size 6353
training_loss.png ADDED