dspfac commited on
Commit
d544143
·
verified ·
1 Parent(s): a8fb713

Upload folder using huggingface_hub

Browse files
Files changed (4) hide show
  1. config.json +102 -0
  2. model.safetensors +3 -0
  3. tokenizer.json +0 -0
  4. tokenizer_config.json +240 -0
config.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openbmb/MiniCPM-SALA",
3
+ "architectures": [
4
+ "MiniCPMSALAForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_use_output_gate": true,
9
+ "attn_use_rope": false,
10
+ "auto_map": {
11
+ "AutoConfig": "configuration_minicpm_sala.MiniCPMSALAConfig",
12
+ "AutoModel": "modeling_minicpm_sala.MiniCPMSALAModel",
13
+ "AutoModelForCausalLM": "modeling_minicpm_sala.MiniCPMSALAForCausalLM",
14
+ "AutoModelForSeq2SeqLM": "modeling_minicpm_sala.MiniCPMSALAForCausalLM",
15
+ "AutoModelForSequenceClassification": "modeling_minicpm_sala.MiniCPMSALAForSequenceClassification"
16
+ },
17
+ "bos_token_id": 1,
18
+ "dim_model_base": 256,
19
+ "dtype": "bfloat16",
20
+ "eos_token_id": [
21
+ 2,
22
+ 73440
23
+ ],
24
+ "head_dim": 128,
25
+ "hidden_act": "silu",
26
+ "hidden_size": 4096,
27
+ "initializer_range": 0.1,
28
+ "intermediate_size": 16384,
29
+ "lightning_head_dim": 128,
30
+ "lightning_nh": 32,
31
+ "lightning_nkv": 32,
32
+ "lightning_scale": "1/sqrt(d)",
33
+ "lightning_use_rope": true,
34
+ "max_position_embeddings": 524288,
35
+ "mixer_types": [
36
+ "minicpm4",
37
+ "lightning-attn",
38
+ "lightning-attn",
39
+ "lightning-attn",
40
+ "lightning-attn",
41
+ "lightning-attn",
42
+ "lightning-attn",
43
+ "lightning-attn",
44
+ "lightning-attn",
45
+ "minicpm4",
46
+ "lightning-attn",
47
+ "lightning-attn",
48
+ "lightning-attn",
49
+ "lightning-attn",
50
+ "lightning-attn",
51
+ "lightning-attn",
52
+ "minicpm4",
53
+ "minicpm4",
54
+ "lightning-attn",
55
+ "lightning-attn",
56
+ "lightning-attn",
57
+ "lightning-attn",
58
+ "minicpm4",
59
+ "lightning-attn",
60
+ "lightning-attn",
61
+ "lightning-attn",
62
+ "lightning-attn",
63
+ "lightning-attn",
64
+ "lightning-attn",
65
+ "minicpm4",
66
+ "minicpm4",
67
+ "minicpm4"
68
+ ],
69
+ "model_type": "minicpm_sala",
70
+ "mup_denominator": 32,
71
+ "num_attention_heads": 32,
72
+ "num_hidden_layers": 32,
73
+ "num_key_value_heads": 2,
74
+ "pad_token_id": 2,
75
+ "qk_norm": true,
76
+ "quantization": {
77
+ "bits": 8,
78
+ "group_size": 64
79
+ },
80
+ "rand_init": false,
81
+ "rms_norm_eps": 1e-6,
82
+ "rope_theta": 10000.0,
83
+ "scale_depth": 1.4,
84
+ "scale_emb": 12,
85
+ "sparse_config": {
86
+ "block_size": 64,
87
+ "dense_len": 8192,
88
+ "init_blocks": 1,
89
+ "kernel_size": 32,
90
+ "kernel_stride": 16,
91
+ "topk": 64,
92
+ "use_nope": false,
93
+ "window_size": 2048
94
+ },
95
+ "tie_word_embeddings": false,
96
+ "torch_dtype": "bfloat16",
97
+ "transformers_version": "4.56.0",
98
+ "use_cache": true,
99
+ "use_output_gate": true,
100
+ "use_output_norm": true,
101
+ "vocab_size": 73448
102
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a35688f5eaee7e3bc5b0e10e8eba4daf08ce8991cec7e7a4d1bf02e05df7f5b0
3
+ size 10352021208
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "101": {
30
+ "content": "<think>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": false
36
+ },
37
+ "102": {
38
+ "content": "</think>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": false
44
+ },
45
+ "103": {
46
+ "content": "<tool_response>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "104": {
54
+ "content": "</tool_response>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "105": {
62
+ "content": "<tools>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "106": {
70
+ "content": "</tools>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "107": {
78
+ "content": "<parameters>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "108": {
86
+ "content": "</parameters>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "109": {
94
+ "content": "<arguments>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "110": {
102
+ "content": "</arguments>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "111": {
110
+ "content": "<function",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "112": {
118
+ "content": "</function>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": true
124
+ },
125
+ "113": {
126
+ "content": "<param",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": true
132
+ },
133
+ "114": {
134
+ "content": "</param>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": true
140
+ },
141
+ "73440": {
142
+ "content": "<|im_end|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": true
148
+ },
149
+ "73441": {
150
+ "content": "<|im_start|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": true
156
+ },
157
+ "73442": {
158
+ "content": "<tool_call>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": true
164
+ },
165
+ "73443": {
166
+ "content": "</tool_call>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": true
172
+ },
173
+ "73444": {
174
+ "content": "<|im_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": true
180
+ },
181
+ "73445": {
182
+ "content": "<|fim_prefix|>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": true
188
+ },
189
+ "73446": {
190
+ "content": "<|fim_middle|>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": true
196
+ },
197
+ "73447": {
198
+ "content": "<|fim_suffix|>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": true
204
+ }
205
+ },
206
+ "additional_special_tokens": [
207
+ "<|im_end|>",
208
+ "<|im_start|>",
209
+ "<tool_call>",
210
+ "</tool_call>",
211
+ "<|im_sep|>",
212
+ "<|fim_prefix|>",
213
+ "<|fim_middle|>",
214
+ "<|fim_suffix|>",
215
+ "<tool_response>",
216
+ "</tool_response>",
217
+ "<tools>",
218
+ "</tools>",
219
+ "<arguments>",
220
+ "</arguments>",
221
+ "<parameters>",
222
+ "</parameters>",
223
+ "<function",
224
+ "</function>",
225
+ "<param",
226
+ "</param>"
227
+ ],
228
+ "bos_token": "<s>",
229
+ "clean_up_tokenization_spaces": false,
230
+ "eos_token": "<|im_end|>",
231
+ "legacy": true,
232
+ "model_max_length": 262144,
233
+ "pad_token": "</s>",
234
+ "sp_model_kwargs": {},
235
+ "spaces_between_special_tokens": false,
236
+ "tokenizer_class": "LlamaTokenizer",
237
+ "unk_token": "<unk>",
238
+ "use_default_system_prompt": false,
239
+ "chat_template": "{%- if tools %}\n {%- set tool_definitions %}\n {{- \"# Tools\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson(ensure_ascii=False) }}\n {%- endfor %}\n {{- '\\n</tools>\\n\\nTool usage guidelines:\\n- You may call zero or more functions. If no function calls are needed, just answer normally and do not include any <function ... </function>.\\n- When calling a function, return an XML object within <function ... </function> using:\\n<function name=\"function-name\"><param name=\"param-name\">param-value</param></function>\\n- param-value may be multi-line. If it contains <, & or newline characters, wrap it in a CDATA block: <param name=\"param-name\"><![CDATA[...multi-line value...]]></param>' }}\n {%- endset %}\n \n {{- '<|im_start|>system\\n' }}\n {%- if messages[0].role == 'system' %}\n {%- if '<tool_def_sep>' in messages[0].content %}\n {{- messages[0].content.replace('<tool_def_sep>', tool_definitions) }}\n {%- else %}\n {{- messages[0].content + '\\n\\n' + tool_definitions }}\n {%- endif %}\n {%- else %}\n {{- tool_definitions.lstrip() }}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n {%- set index = (messages|length - 1) - loop.index0 %}\n {%- 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>')) %}\n {%- set ns.multi_step_tool = false %}\n {%- set ns.last_query_index = index %}\n {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n {%- if message.content is string %}\n {%- set content = message.content %}\n {%- else %}\n {%- set content = '' %}\n {%- endif %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {%- set reasoning_content = '' %}\n {%- if message.reasoning_content is string %}\n {%- set reasoning_content = message.reasoning_content %}\n {%- else %}\n {%- if '</think>' in content %}\n {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n {%- endif %}\n {%- endif %}\n \n {%- if message.tool_calls %}\n {%- set content_parts = content.split('<tool_sep>') %}\n {%- set processed_content = content_parts[0] %}\n {%- set tool_calls_count = message.tool_calls|length %}\n {%- set tool_sep_count = content_parts|length - 1 %}\n {%- set min_count = [tool_calls_count, tool_sep_count]|min %}\n \n {%- for i in range(1, content_parts|length) %}\n {%- set tool_index = i - 1 %}\n {%- if tool_index < tool_calls_count %}\n {%- set tool_call = message.tool_calls[tool_index] %}\n {%- if tool_call.function %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {%- set single_tool_xml %}\n {{- '<function name=\"' ~ tool_call.name ~ '\">' }}\n {%- if tool_call.arguments %}\n {%- set args_dict = tool_call.arguments %}\n {%- for param_name, param_value in args_dict.items() %}\n {{- '<param name=\"' ~ param_name ~ '\">' }}\n {%- if param_value is string and ('<' in param_value or '&' in param_value or '\\n' in param_value) %}\n {{- '<![CDATA[' + param_value + ']]>' }}\n {%- else %}\n {{- param_value }}\n {%- endif %}\n {{- '</param>' }}\n {%- endfor %}\n {%- endif %}\n {{- '</function>' }}\n {%- endset %}\n {%- set processed_content = processed_content + single_tool_xml + content_parts[i] %}\n {%- else %}\n {%- set processed_content = processed_content + content_parts[i] %}\n {%- endif %}\n {%- endfor %}\n \n {%- if tool_calls_count > tool_sep_count %}\n {%- for remaining_index in range(tool_sep_count, tool_calls_count) %}\n {%- set tool_call = message.tool_calls[remaining_index] %}\n {%- if tool_call.function %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {%- set remaining_tool_xml %}\n {{- '<function name=\"' ~ tool_call.name ~ '\">' }}\n {%- if tool_call.arguments %}\n {%- set args_dict = tool_call.arguments %}\n {%- for param_name, param_value in args_dict.items() %}\n {{- '<param name=\"' ~ param_name ~ '\">' }}\n {%- if param_value is string and ('<' in param_value or '&' in param_value or '\\n' in param_value) %}\n {{- '<![CDATA[' + param_value + ']]>' }}\n {%- else %}\n {{- param_value }}\n {%- endif %}\n {{- '</param>' }}\n {%- endfor %}\n {%- endif %}\n {{- '</function>' }}\n {%- endset %}\n {%- set processed_content = processed_content + remaining_tool_xml %}\n {%- endfor %}\n {%- endif %}\n \n {%- set content = processed_content %}\n {%- endif %}\n \n {%- if loop.index0 > ns.last_query_index %}\n {%- if reasoning_content %}\n {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n \n {%- if message.tool_calls and not has_tool_sep %}\n {%- for tool_call in message.tool_calls %}\n {%- if (loop.first and content) or (not loop.first) %}\n {{- '\\n' }}\n {%- endif %}\n {%- if tool_call.function %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '<function name=\"' ~ tool_call.name ~ '\">' }}\n {%- if tool_call.arguments %}\n {%- set args_dict = tool_call.arguments %}\n {%- for param_name, param_value in args_dict.items() %}\n {{- '<param name=\"' ~ param_name ~ '\">' }}\n {%- if param_value is string and ('<' in param_value or '&' in param_value or '\\n' in param_value) %}\n {{- '<![CDATA[' + param_value + ']]>' }}\n {%- else %}\n {{- param_value }}\n {%- endif %}\n {{- '</param>' }}\n {%- endfor %}\n {%- endif %}\n {{- '</function>' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {%- if message.content is string %}\n {{- content }}\n {%- else %}\n {{- message.content | tojson(ensure_ascii=False) }}\n {%- endif %}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}"
240
+ }