Hugging Face's logo

Lorbus
/
GLM52MTPLayer78

Safetensors
glm_moe_dsa
Model card Files
xet
 Community
Lorbus commited on
Commit
4ed16ce
·
verified ·
1 Parent(s): eb6d456

Upload GLM-5.2 MTP spec model

Browse files
Files changed (8) hide show
  1. .gitattributes +1 -0
  2. chat_template.jinja +119 -0
  3. config.json +223 -0
  4. generation_config.json +12 -0
  5. model.safetensors +3 -0
  6. model.safetensors.index.json +798 -0
  7. tokenizer.json +3 -0
  8. tokenizer_config.json +33 -0
.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
chat_template.jinja ADDED
@@ -0,0 +1,119 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [gMASK]<sop>
2
+ {%- set effective_reasoning_effort = 'high' if reasoning_effort is defined and reasoning_effort == 'high' else 'max' -%}
3
+ {%- if (enable_thinking is not defined or enable_thinking) and effective_reasoning_effort is not none -%}<|system|>Reasoning Effort: {{ effective_reasoning_effort | capitalize }}{%- endif -%}
4
+ {%- if tools -%}
5
+ {%- macro tool_to_json(tool) -%}
6
+ {%- set ns_tool = namespace(first=true) -%}
7
+ {{ '{' -}}
8
+ {%- for k, v in tool.items() -%}
9
+ {%- if k != 'defer_loading' and k != 'strict' -%}
10
+ {%- if not ns_tool.first -%}{{- ', ' -}}{%- endif -%}
11
+ {%- set ns_tool.first = false -%}
12
+ "{{ k }}": {{ v | tojson(ensure_ascii=False) }}
13
+ {%- endif -%}
14
+ {%- endfor -%}
15
+ {{- '}' -}}
16
+ {%- endmacro -%}
17
+ <|system|>
18
+ # Tools
19
+
20
+ You may call one or more functions to assist with the user query.
21
+
22
+ You are provided with function signatures within <tools></tools> XML tags:
23
+ <tools>
24
+ {% for tool in tools %}
25
+ {%- if 'function' in tool -%}
26
+ {%- set tool = tool['function'] -%}
27
+ {%- endif -%}
28
+ {% if tool.defer_loading is not defined or not tool.defer_loading %}
29
+ {{ tool_to_json(tool) }}
30
+ {% endif %}
31
+ {% endfor %}
32
+ </tools>
33
+
34
+ For each function call, output the function name and arguments within the following XML format:
35
+ <tool_call>{function-name}<arg_key>{arg-key-1}</arg_key><arg_value>{arg-value-1}</arg_value><arg_key>{arg-key-2}</arg_key><arg_value>{arg-value-2}</arg_value>...</tool_call>{%- endif -%}
36
+ {%- macro visible_text(content) -%}
37
+ {%- if content is string -%}
38
+ {{- content }}
39
+ {%- elif content is iterable and content is not mapping -%}
40
+ {%- for item in content -%}
41
+ {%- if item is mapping and item.type == 'text' -%}
42
+ {{- item.text }}
43
+ {%- elif item is string -%}
44
+ {{- item }}
45
+ {%- elif item is mapping and item.type in ['image', 'image_url', 'video', 'video_url', 'audio', 'audio_url', 'input_audio'] -%}
46
+ {%- set media_type = item.type | replace('_url', '') | replace('input_', '') -%}
47
+ {{- "<reminder>You are unable to process this " ~ media_type ~ " because you don't have multi-modal input ability. Try different methods.</reminder>" }}
48
+ {%- endif -%}
49
+ {%- endfor -%}
50
+ {%- else -%}
51
+ {{- content }}
52
+ {%- endif -%}
53
+ {%- endmacro -%}
54
+ {%- set ns = namespace(last_user_index=-1) -%}
55
+ {%- for m in messages %}
56
+ {%- if m.role == 'user' %}
57
+ {%- set ns.last_user_index = loop.index0 -%}
58
+ {%- endif %}
59
+ {%- endfor %}
60
+ {%- for m in messages -%}
61
+ {%- if m.role == 'user' -%}<|user|>{{ visible_text(m.content) }}
62
+ {%- elif m.role == 'assistant' -%}
63
+ <|assistant|>
64
+ {%- set content = visible_text(m.content) %}
65
+ {%- if m.reasoning_content is string %}
66
+ {%- set reasoning_content = m.reasoning_content %}
67
+ {%- elif '</think>' in content %}
68
+ {%- set reasoning_content = content.split('</think>')[0].split('<think>')[-1] %}
69
+ {%- set content = content.split('</think>')[-1] %}
70
+ {%- endif %}
71
+ {%- if ((clear_thinking is defined and not clear_thinking) or loop.index0 > ns.last_user_index) and reasoning_content is defined -%}
72
+ {{ '<think>' + reasoning_content + '</think>'}}
73
+ {%- else -%}
74
+ {{ '<think></think>' }}
75
+ {%- endif -%}
76
+ {%- if content.strip() -%}
77
+ {{ content.strip() }}
78
+ {%- endif -%}
79
+ {% if m.tool_calls %}
80
+ {% for tc in m.tool_calls %}
81
+ {%- if tc.function %}
82
+ {%- set tc = tc.function %}
83
+ {%- endif %}
84
+ {{- '<tool_call>' + tc.name -}}
85
+ {% set _args = tc.arguments %}{% for k, v in _args.items() %}<arg_key>{{ k }}</arg_key><arg_value>{{ v | tojson(ensure_ascii=False) if v is not string else v }}</arg_value>{% endfor %}</tool_call>{% endfor %}
86
+ {% endif %}
87
+ {%- elif m.role == 'tool' -%}
88
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
89
+ {{- '<|observation|>' -}}
90
+ {%- endif %}
91
+ {%- if m.content is string -%}
92
+ {{- '<tool_response>' + m.content + '</tool_response>' -}}
93
+ {%- elif m.content is iterable and m.content is not mapping and m.content and m.content.0.type == "tool_reference" -%}
94
+ {{- '<tool_response><tools>\n' -}}
95
+ {% for tr in m.content %}
96
+ {%- for tool in tools -%}
97
+ {%- if 'function' in tool -%}
98
+ {%- set tool = tool['function'] -%}
99
+ {%- endif -%}
100
+ {%- if tool.name == tr.name -%}
101
+ {{- tool_to_json(tool) + '\n' -}}
102
+ {%- endif -%}
103
+ {%- endfor -%}
104
+ {%- endfor -%}
105
+ {{- '</tools></tool_response>' -}}
106
+ {%- elif m.content is iterable and m.content is not mapping and m.content and m.content.0 is mapping and m.content.0.output is defined -%}
107
+ {%- for tr in m.content -%}
108
+ {{- '<tool_response>' + tr.output + '</tool_response>' -}}
109
+ {%- endfor -%}
110
+ {%- else -%}
111
+ {{- '<tool_response>' + visible_text(m.content) + '</tool_response>' -}}
112
+ {% endif -%}
113
+ {%- elif m.role == 'system' -%}
114
+ <|system|>{{ visible_text(m.content) }}
115
+ {%- endif -%}
116
+ {%- endfor -%}
117
+ {%- if add_generation_prompt -%}
118
+ <|assistant|>{{- '<think></think>' if (enable_thinking is defined and not enable_thinking) else '<think>' -}}
119
+ {%- endif -%}
config.json ADDED
@@ -0,0 +1,223 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "GlmMoeDsaForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "dtype": "bfloat16",
8
+ "eos_token_id": [
9
+ 154820,
10
+ 154827,
11
+ 154829
12
+ ],
13
+ "ep_size": 1,
14
+ "first_k_dense_replace": 3,
15
+ "head_dim": 192,
16
+ "hidden_act": "silu",
17
+ "hidden_size": 6144,
18
+ "index_head_dim": 128,
19
+ "index_n_heads": 32,
20
+ "index_share_for_mtp_iteration": true,
21
+ "index_skip_topk_offset": 3,
22
+ "index_topk": 2048,
23
+ "index_topk_freq": 4,
24
+ "index_topk_pattern": null,
25
+ "indexer_rope_interleave": true,
26
+ "indexer_types": [
27
+ "full",
28
+ "full",
29
+ "full",
30
+ "shared",
31
+ "shared",
32
+ "shared",
33
+ "full",
34
+ "shared",
35
+ "shared",
36
+ "shared",
37
+ "full",
38
+ "shared",
39
+ "shared",
40
+ "shared",
41
+ "full",
42
+ "shared",
43
+ "shared",
44
+ "shared",
45
+ "full",
46
+ "shared",
47
+ "shared",
48
+ "shared",
49
+ "full",
50
+ "shared",
51
+ "shared",
52
+ "shared",
53
+ "full",
54
+ "shared",
55
+ "shared",
56
+ "shared",
57
+ "full",
58
+ "shared",
59
+ "shared",
60
+ "shared",
61
+ "full",
62
+ "shared",
63
+ "shared",
64
+ "shared",
65
+ "full",
66
+ "shared",
67
+ "shared",
68
+ "shared",
69
+ "full",
70
+ "shared",
71
+ "shared",
72
+ "shared",
73
+ "full",
74
+ "shared",
75
+ "shared",
76
+ "shared",
77
+ "full",
78
+ "shared",
79
+ "shared",
80
+ "shared",
81
+ "full",
82
+ "shared",
83
+ "shared",
84
+ "shared",
85
+ "full",
86
+ "shared",
87
+ "shared",
88
+ "shared",
89
+ "full",
90
+ "shared",
91
+ "shared",
92
+ "shared",
93
+ "full",
94
+ "shared",
95
+ "shared",
96
+ "shared",
97
+ "full",
98
+ "shared",
99
+ "shared",
100
+ "shared",
101
+ "full",
102
+ "shared",
103
+ "shared",
104
+ "shared"
105
+ ],
106
+ "initializer_range": 0.02,
107
+ "intermediate_size": 12288,
108
+ "kv_lora_rank": 512,
109
+ "max_position_embeddings": 1048576,
110
+ "mlp_layer_types": [
111
+ "dense",
112
+ "dense",
113
+ "dense",
114
+ "sparse",
115
+ "sparse",
116
+ "sparse",
117
+ "sparse",
118
+ "sparse",
119
+ "sparse",
120
+ "sparse",
121
+ "sparse",
122
+ "sparse",
123
+ "sparse",
124
+ "sparse",
125
+ "sparse",
126
+ "sparse",
127
+ "sparse",
128
+ "sparse",
129
+ "sparse",
130
+ "sparse",
131
+ "sparse",
132
+ "sparse",
133
+ "sparse",
134
+ "sparse",
135
+ "sparse",
136
+ "sparse",
137
+ "sparse",
138
+ "sparse",
139
+ "sparse",
140
+ "sparse",
141
+ "sparse",
142
+ "sparse",
143
+ "sparse",
144
+ "sparse",
145
+ "sparse",
146
+ "sparse",
147
+ "sparse",
148
+ "sparse",
149
+ "sparse",
150
+ "sparse",
151
+ "sparse",
152
+ "sparse",
153
+ "sparse",
154
+ "sparse",
155
+ "sparse",
156
+ "sparse",
157
+ "sparse",
158
+ "sparse",
159
+ "sparse",
160
+ "sparse",
161
+ "sparse",
162
+ "sparse",
163
+ "sparse",
164
+ "sparse",
165
+ "sparse",
166
+ "sparse",
167
+ "sparse",
168
+ "sparse",
169
+ "sparse",
170
+ "sparse",
171
+ "sparse",
172
+ "sparse",
173
+ "sparse",
174
+ "sparse",
175
+ "sparse",
176
+ "sparse",
177
+ "sparse",
178
+ "sparse",
179
+ "sparse",
180
+ "sparse",
181
+ "sparse",
182
+ "sparse",
183
+ "sparse",
184
+ "sparse",
185
+ "sparse",
186
+ "sparse",
187
+ "sparse",
188
+ "sparse"
189
+ ],
190
+ "model_type": "glm_moe_dsa",
191
+ "moe_intermediate_size": 2048,
192
+ "moe_layer_freq": 1,
193
+ "n_group": 1,
194
+ "n_routed_experts": 256,
195
+ "n_shared_experts": 1,
196
+ "norm_topk_prob": true,
197
+ "num_attention_heads": 64,
198
+ "num_experts_per_tok": 8,
199
+ "num_hidden_layers": 78,
200
+ "num_key_value_heads": 64,
201
+ "num_nextn_predict_layers": 1,
202
+ "pad_token_id": 154820,
203
+ "pretraining_tp": 1,
204
+ "q_lora_rank": 2048,
205
+ "qk_head_dim": 256,
206
+ "qk_nope_head_dim": 192,
207
+ "qk_rope_head_dim": 64,
208
+ "rms_norm_eps": 1e-05,
209
+ "rope_interleave": true,
210
+ "rope_parameters": {
211
+ "rope_theta": 8000000,
212
+ "rope_type": "default"
213
+ },
214
+ "routed_scaling_factor": 2.5,
215
+ "scoring_func": "sigmoid",
216
+ "tie_word_embeddings": false,
217
+ "topk_group": 1,
218
+ "topk_method": "noaux_tc",
219
+ "transformers_version": "5.12.0",
220
+ "use_cache": true,
221
+ "v_head_dim": 256,
222
+ "vocab_size": 154880
223
+ }
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "eos_token_id": [
4
+ 154820,
5
+ 154827,
6
+ 154829
7
+ ],
8
+ "pad_token_id": 154820,
9
+ "temperature": 1.0,
10
+ "top_p": 0.95,
11
+ "transformers_version": "5.12.0"
12
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18a22cc4148cdb11e91acecfc1a7524e9cabbf6895b7117f477289ebef18a37e
3
+ size 19906347288
model.safetensors.index.json ADDED
@@ -0,0 +1,798 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 19906247168
4
+ },
5
+ "weight_map": {
6
+ "model.layers.78.eh_proj.weight": "model.safetensors",
7
+ "model.layers.78.enorm.weight": "model.safetensors",
8
+ "model.layers.78.hnorm.weight": "model.safetensors",
9
+ "model.layers.78.input_layernorm.weight": "model.safetensors",
10
+ "model.layers.78.mlp.experts.0.down_proj.weight": "model.safetensors",
11
+ "model.layers.78.mlp.experts.0.gate_proj.weight": "model.safetensors",
12
+ "model.layers.78.mlp.experts.0.up_proj.weight": "model.safetensors",
13
+ "model.layers.78.mlp.experts.1.down_proj.weight": "model.safetensors",
14
+ "model.layers.78.mlp.experts.1.gate_proj.weight": "model.safetensors",
15
+ "model.layers.78.mlp.experts.1.up_proj.weight": "model.safetensors",
16
+ "model.layers.78.mlp.experts.10.down_proj.weight": "model.safetensors",
17
+ "model.layers.78.mlp.experts.10.gate_proj.weight": "model.safetensors",
18
+ "model.layers.78.mlp.experts.10.up_proj.weight": "model.safetensors",
19
+ "model.layers.78.mlp.experts.100.down_proj.weight": "model.safetensors",
20
+ "model.layers.78.mlp.experts.100.gate_proj.weight": "model.safetensors",
21
+ "model.layers.78.mlp.experts.100.up_proj.weight": "model.safetensors",
22
+ "model.layers.78.mlp.experts.101.down_proj.weight": "model.safetensors",
23
+ "model.layers.78.mlp.experts.101.gate_proj.weight": "model.safetensors",
24
+ "model.layers.78.mlp.experts.101.up_proj.weight": "model.safetensors",
25
+ "model.layers.78.mlp.experts.102.down_proj.weight": "model.safetensors",
26
+ "model.layers.78.mlp.experts.102.gate_proj.weight": "model.safetensors",
27
+ "model.layers.78.mlp.experts.102.up_proj.weight": "model.safetensors",
28
+ "model.layers.78.mlp.experts.103.down_proj.weight": "model.safetensors",
29
+ "model.layers.78.mlp.experts.103.gate_proj.weight": "model.safetensors",
30
+ "model.layers.78.mlp.experts.103.up_proj.weight": "model.safetensors",
31
+ "model.layers.78.mlp.experts.104.down_proj.weight": "model.safetensors",
32
+ "model.layers.78.mlp.experts.104.gate_proj.weight": "model.safetensors",
33
+ "model.layers.78.mlp.experts.104.up_proj.weight": "model.safetensors",
34
+ "model.layers.78.mlp.experts.105.down_proj.weight": "model.safetensors",
35
+ "model.layers.78.mlp.experts.105.gate_proj.weight": "model.safetensors",
36
+ "model.layers.78.mlp.experts.105.up_proj.weight": "model.safetensors",
37
+ "model.layers.78.mlp.experts.106.down_proj.weight": "model.safetensors",
38
+ "model.layers.78.mlp.experts.106.gate_proj.weight": "model.safetensors",
39
+ "model.layers.78.mlp.experts.106.up_proj.weight": "model.safetensors",
40
+ "model.layers.78.mlp.experts.107.down_proj.weight": "model.safetensors",
41
+ "model.layers.78.mlp.experts.107.gate_proj.weight": "model.safetensors",
42
+ "model.layers.78.mlp.experts.107.up_proj.weight": "model.safetensors",
43
+ "model.layers.78.mlp.experts.108.down_proj.weight": "model.safetensors",
44
+ "model.layers.78.mlp.experts.108.gate_proj.weight": "model.safetensors",
45
+ "model.layers.78.mlp.experts.108.up_proj.weight": "model.safetensors",
46
+ "model.layers.78.mlp.experts.109.down_proj.weight": "model.safetensors",
47
+ "model.layers.78.mlp.experts.109.gate_proj.weight": "model.safetensors",
48
+ "model.layers.78.mlp.experts.109.up_proj.weight": "model.safetensors",
49
+ "model.layers.78.mlp.experts.11.down_proj.weight": "model.safetensors",
50
+ "model.layers.78.mlp.experts.11.gate_proj.weight": "model.safetensors",
51
+ "model.layers.78.mlp.experts.11.up_proj.weight": "model.safetensors",
52
+ "model.layers.78.mlp.experts.110.down_proj.weight": "model.safetensors",
53
+ "model.layers.78.mlp.experts.110.gate_proj.weight": "model.safetensors",
54
+ "model.layers.78.mlp.experts.110.up_proj.weight": "model.safetensors",
55
+ "model.layers.78.mlp.experts.111.down_proj.weight": "model.safetensors",
56
+ "model.layers.78.mlp.experts.111.gate_proj.weight": "model.safetensors",
57
+ "model.layers.78.mlp.experts.111.up_proj.weight": "model.safetensors",
58
+ "model.layers.78.mlp.experts.112.down_proj.weight": "model.safetensors",
59
+ "model.layers.78.mlp.experts.112.gate_proj.weight": "model.safetensors",
60
+ "model.layers.78.mlp.experts.112.up_proj.weight": "model.safetensors",
61
+ "model.layers.78.mlp.experts.113.down_proj.weight": "model.safetensors",
62
+ "model.layers.78.mlp.experts.113.gate_proj.weight": "model.safetensors",
63
+ "model.layers.78.mlp.experts.113.up_proj.weight": "model.safetensors",
64
+ "model.layers.78.mlp.experts.114.down_proj.weight": "model.safetensors",
65
+ "model.layers.78.mlp.experts.114.gate_proj.weight": "model.safetensors",
66
+ "model.layers.78.mlp.experts.114.up_proj.weight": "model.safetensors",
67
+ "model.layers.78.mlp.experts.115.down_proj.weight": "model.safetensors",
68
+ "model.layers.78.mlp.experts.115.gate_proj.weight": "model.safetensors",
69
+ "model.layers.78.mlp.experts.115.up_proj.weight": "model.safetensors",
70
+ "model.layers.78.mlp.experts.116.down_proj.weight": "model.safetensors",
71
+ "model.layers.78.mlp.experts.116.gate_proj.weight": "model.safetensors",
72
+ "model.layers.78.mlp.experts.116.up_proj.weight": "model.safetensors",
73
+ "model.layers.78.mlp.experts.117.down_proj.weight": "model.safetensors",
74
+ "model.layers.78.mlp.experts.117.gate_proj.weight": "model.safetensors",
75
+ "model.layers.78.mlp.experts.117.up_proj.weight": "model.safetensors",
76
+ "model.layers.78.mlp.experts.118.down_proj.weight": "model.safetensors",
77
+ "model.layers.78.mlp.experts.118.gate_proj.weight": "model.safetensors",
78
+ "model.layers.78.mlp.experts.118.up_proj.weight": "model.safetensors",
79
+ "model.layers.78.mlp.experts.119.down_proj.weight": "model.safetensors",
80
+ "model.layers.78.mlp.experts.119.gate_proj.weight": "model.safetensors",
81
+ "model.layers.78.mlp.experts.119.up_proj.weight": "model.safetensors",
82
+ "model.layers.78.mlp.experts.12.down_proj.weight": "model.safetensors",
83
+ "model.layers.78.mlp.experts.12.gate_proj.weight": "model.safetensors",
84
+ "model.layers.78.mlp.experts.12.up_proj.weight": "model.safetensors",
85
+ "model.layers.78.mlp.experts.120.down_proj.weight": "model.safetensors",
86
+ "model.layers.78.mlp.experts.120.gate_proj.weight": "model.safetensors",
87
+ "model.layers.78.mlp.experts.120.up_proj.weight": "model.safetensors",
88
+ "model.layers.78.mlp.experts.121.down_proj.weight": "model.safetensors",
89
+ "model.layers.78.mlp.experts.121.gate_proj.weight": "model.safetensors",
90
+ "model.layers.78.mlp.experts.121.up_proj.weight": "model.safetensors",
91
+ "model.layers.78.mlp.experts.122.down_proj.weight": "model.safetensors",
92
+ "model.layers.78.mlp.experts.122.gate_proj.weight": "model.safetensors",
93
+ "model.layers.78.mlp.experts.122.up_proj.weight": "model.safetensors",
94
+ "model.layers.78.mlp.experts.123.down_proj.weight": "model.safetensors",
95
+ "model.layers.78.mlp.experts.123.gate_proj.weight": "model.safetensors",
96
+ "model.layers.78.mlp.experts.123.up_proj.weight": "model.safetensors",
97
+ "model.layers.78.mlp.experts.124.down_proj.weight": "model.safetensors",
98
+ "model.layers.78.mlp.experts.124.gate_proj.weight": "model.safetensors",
99
+ "model.layers.78.mlp.experts.124.up_proj.weight": "model.safetensors",
100
+ "model.layers.78.mlp.experts.125.down_proj.weight": "model.safetensors",
101
+ "model.layers.78.mlp.experts.125.gate_proj.weight": "model.safetensors",
102
+ "model.layers.78.mlp.experts.125.up_proj.weight": "model.safetensors",
103
+ "model.layers.78.mlp.experts.126.down_proj.weight": "model.safetensors",
104
+ "model.layers.78.mlp.experts.126.gate_proj.weight": "model.safetensors",
105
+ "model.layers.78.mlp.experts.126.up_proj.weight": "model.safetensors",
106
+ "model.layers.78.mlp.experts.127.down_proj.weight": "model.safetensors",
107
+ "model.layers.78.mlp.experts.127.gate_proj.weight": "model.safetensors",
108
+ "model.layers.78.mlp.experts.127.up_proj.weight": "model.safetensors",
109
+ "model.layers.78.mlp.experts.128.down_proj.weight": "model.safetensors",
110
+ "model.layers.78.mlp.experts.128.gate_proj.weight": "model.safetensors",
111
+ "model.layers.78.mlp.experts.128.up_proj.weight": "model.safetensors",
112
+ "model.layers.78.mlp.experts.129.down_proj.weight": "model.safetensors",
113
+ "model.layers.78.mlp.experts.129.gate_proj.weight": "model.safetensors",
114
+ "model.layers.78.mlp.experts.129.up_proj.weight": "model.safetensors",
115
+ "model.layers.78.mlp.experts.13.down_proj.weight": "model.safetensors",
116
+ "model.layers.78.mlp.experts.13.gate_proj.weight": "model.safetensors",
117
+ "model.layers.78.mlp.experts.13.up_proj.weight": "model.safetensors",
118
+ "model.layers.78.mlp.experts.130.down_proj.weight": "model.safetensors",
119
+ "model.layers.78.mlp.experts.130.gate_proj.weight": "model.safetensors",
120
+ "model.layers.78.mlp.experts.130.up_proj.weight": "model.safetensors",
121
+ "model.layers.78.mlp.experts.131.down_proj.weight": "model.safetensors",
122
+ "model.layers.78.mlp.experts.131.gate_proj.weight": "model.safetensors",
123
+ "model.layers.78.mlp.experts.131.up_proj.weight": "model.safetensors",
124
+ "model.layers.78.mlp.experts.132.down_proj.weight": "model.safetensors",
125
+ "model.layers.78.mlp.experts.132.gate_proj.weight": "model.safetensors",
126
+ "model.layers.78.mlp.experts.132.up_proj.weight": "model.safetensors",
127
+ "model.layers.78.mlp.experts.133.down_proj.weight": "model.safetensors",
128
+ "model.layers.78.mlp.experts.133.gate_proj.weight": "model.safetensors",
129
+ "model.layers.78.mlp.experts.133.up_proj.weight": "model.safetensors",
130
+ "model.layers.78.mlp.experts.134.down_proj.weight": "model.safetensors",
131
+ "model.layers.78.mlp.experts.134.gate_proj.weight": "model.safetensors",
132
+ "model.layers.78.mlp.experts.134.up_proj.weight": "model.safetensors",
133
+ "model.layers.78.mlp.experts.135.down_proj.weight": "model.safetensors",
134
+ "model.layers.78.mlp.experts.135.gate_proj.weight": "model.safetensors",
135
+ "model.layers.78.mlp.experts.135.up_proj.weight": "model.safetensors",
136
+ "model.layers.78.mlp.experts.136.down_proj.weight": "model.safetensors",
137
+ "model.layers.78.mlp.experts.136.gate_proj.weight": "model.safetensors",
138
+ "model.layers.78.mlp.experts.136.up_proj.weight": "model.safetensors",
139
+ "model.layers.78.mlp.experts.137.down_proj.weight": "model.safetensors",
140
+ "model.layers.78.mlp.experts.137.gate_proj.weight": "model.safetensors",
141
+ "model.layers.78.mlp.experts.137.up_proj.weight": "model.safetensors",
142
+ "model.layers.78.mlp.experts.138.down_proj.weight": "model.safetensors",
143
+ "model.layers.78.mlp.experts.138.gate_proj.weight": "model.safetensors",
144
+ "model.layers.78.mlp.experts.138.up_proj.weight": "model.safetensors",
145
+ "model.layers.78.mlp.experts.139.down_proj.weight": "model.safetensors",
146
+ "model.layers.78.mlp.experts.139.gate_proj.weight": "model.safetensors",
147
+ "model.layers.78.mlp.experts.139.up_proj.weight": "model.safetensors",
148
+ "model.layers.78.mlp.experts.14.down_proj.weight": "model.safetensors",
149
+ "model.layers.78.mlp.experts.14.gate_proj.weight": "model.safetensors",
150
+ "model.layers.78.mlp.experts.14.up_proj.weight": "model.safetensors",
151
+ "model.layers.78.mlp.experts.140.down_proj.weight": "model.safetensors",
152
+ "model.layers.78.mlp.experts.140.gate_proj.weight": "model.safetensors",
153
+ "model.layers.78.mlp.experts.140.up_proj.weight": "model.safetensors",
154
+ "model.layers.78.mlp.experts.141.down_proj.weight": "model.safetensors",
155
+ "model.layers.78.mlp.experts.141.gate_proj.weight": "model.safetensors",
156
+ "model.layers.78.mlp.experts.141.up_proj.weight": "model.safetensors",
157
+ "model.layers.78.mlp.experts.142.down_proj.weight": "model.safetensors",
158
+ "model.layers.78.mlp.experts.142.gate_proj.weight": "model.safetensors",
159
+ "model.layers.78.mlp.experts.142.up_proj.weight": "model.safetensors",
160
+ "model.layers.78.mlp.experts.143.down_proj.weight": "model.safetensors",
161
+ "model.layers.78.mlp.experts.143.gate_proj.weight": "model.safetensors",
162
+ "model.layers.78.mlp.experts.143.up_proj.weight": "model.safetensors",
163
+ "model.layers.78.mlp.experts.144.down_proj.weight": "model.safetensors",
164
+ "model.layers.78.mlp.experts.144.gate_proj.weight": "model.safetensors",
165
+ "model.layers.78.mlp.experts.144.up_proj.weight": "model.safetensors",
166
+ "model.layers.78.mlp.experts.145.down_proj.weight": "model.safetensors",
167
+ "model.layers.78.mlp.experts.145.gate_proj.weight": "model.safetensors",
168
+ "model.layers.78.mlp.experts.145.up_proj.weight": "model.safetensors",
169
+ "model.layers.78.mlp.experts.146.down_proj.weight": "model.safetensors",
170
+ "model.layers.78.mlp.experts.146.gate_proj.weight": "model.safetensors",
171
+ "model.layers.78.mlp.experts.146.up_proj.weight": "model.safetensors",
172
+ "model.layers.78.mlp.experts.147.down_proj.weight": "model.safetensors",
173
+ "model.layers.78.mlp.experts.147.gate_proj.weight": "model.safetensors",
174
+ "model.layers.78.mlp.experts.147.up_proj.weight": "model.safetensors",
175
+ "model.layers.78.mlp.experts.148.down_proj.weight": "model.safetensors",
176
+ "model.layers.78.mlp.experts.148.gate_proj.weight": "model.safetensors",
177
+ "model.layers.78.mlp.experts.148.up_proj.weight": "model.safetensors",
178
+ "model.layers.78.mlp.experts.149.down_proj.weight": "model.safetensors",
179
+ "model.layers.78.mlp.experts.149.gate_proj.weight": "model.safetensors",
180
+ "model.layers.78.mlp.experts.149.up_proj.weight": "model.safetensors",
181
+ "model.layers.78.mlp.experts.15.down_proj.weight": "model.safetensors",
182
+ "model.layers.78.mlp.experts.15.gate_proj.weight": "model.safetensors",
183
+ "model.layers.78.mlp.experts.15.up_proj.weight": "model.safetensors",
184
+ "model.layers.78.mlp.experts.150.down_proj.weight": "model.safetensors",
185
+ "model.layers.78.mlp.experts.150.gate_proj.weight": "model.safetensors",
186
+ "model.layers.78.mlp.experts.150.up_proj.weight": "model.safetensors",
187
+ "model.layers.78.mlp.experts.151.down_proj.weight": "model.safetensors",
188
+ "model.layers.78.mlp.experts.151.gate_proj.weight": "model.safetensors",
189
+ "model.layers.78.mlp.experts.151.up_proj.weight": "model.safetensors",
190
+ "model.layers.78.mlp.experts.152.down_proj.weight": "model.safetensors",
191
+ "model.layers.78.mlp.experts.152.gate_proj.weight": "model.safetensors",
192
+ "model.layers.78.mlp.experts.152.up_proj.weight": "model.safetensors",
193
+ "model.layers.78.mlp.experts.153.down_proj.weight": "model.safetensors",
194
+ "model.layers.78.mlp.experts.153.gate_proj.weight": "model.safetensors",
195
+ "model.layers.78.mlp.experts.153.up_proj.weight": "model.safetensors",
196
+ "model.layers.78.mlp.experts.154.down_proj.weight": "model.safetensors",
197
+ "model.layers.78.mlp.experts.154.gate_proj.weight": "model.safetensors",
198
+ "model.layers.78.mlp.experts.154.up_proj.weight": "model.safetensors",
199
+ "model.layers.78.mlp.experts.155.down_proj.weight": "model.safetensors",
200
+ "model.layers.78.mlp.experts.155.gate_proj.weight": "model.safetensors",
201
+ "model.layers.78.mlp.experts.155.up_proj.weight": "model.safetensors",
202
+ "model.layers.78.mlp.experts.156.down_proj.weight": "model.safetensors",
203
+ "model.layers.78.mlp.experts.156.gate_proj.weight": "model.safetensors",
204
+ "model.layers.78.mlp.experts.156.up_proj.weight": "model.safetensors",
205
+ "model.layers.78.mlp.experts.157.down_proj.weight": "model.safetensors",
206
+ "model.layers.78.mlp.experts.157.gate_proj.weight": "model.safetensors",
207
+ "model.layers.78.mlp.experts.157.up_proj.weight": "model.safetensors",
208
+ "model.layers.78.mlp.experts.158.down_proj.weight": "model.safetensors",
209
+ "model.layers.78.mlp.experts.158.gate_proj.weight": "model.safetensors",
210
+ "model.layers.78.mlp.experts.158.up_proj.weight": "model.safetensors",
211
+ "model.layers.78.mlp.experts.159.down_proj.weight": "model.safetensors",
212
+ "model.layers.78.mlp.experts.159.gate_proj.weight": "model.safetensors",
213
+ "model.layers.78.mlp.experts.159.up_proj.weight": "model.safetensors",
214
+ "model.layers.78.mlp.experts.16.down_proj.weight": "model.safetensors",
215
+ "model.layers.78.mlp.experts.16.gate_proj.weight": "model.safetensors",
216
+ "model.layers.78.mlp.experts.16.up_proj.weight": "model.safetensors",
217
+ "model.layers.78.mlp.experts.160.down_proj.weight": "model.safetensors",
218
+ "model.layers.78.mlp.experts.160.gate_proj.weight": "model.safetensors",
219
+ "model.layers.78.mlp.experts.160.up_proj.weight": "model.safetensors",
220
+ "model.layers.78.mlp.experts.161.down_proj.weight": "model.safetensors",
221
+ "model.layers.78.mlp.experts.161.gate_proj.weight": "model.safetensors",
222
+ "model.layers.78.mlp.experts.161.up_proj.weight": "model.safetensors",
223
+ "model.layers.78.mlp.experts.162.down_proj.weight": "model.safetensors",
224
+ "model.layers.78.mlp.experts.162.gate_proj.weight": "model.safetensors",
225
+ "model.layers.78.mlp.experts.162.up_proj.weight": "model.safetensors",
226
+ "model.layers.78.mlp.experts.163.down_proj.weight": "model.safetensors",
227
+ "model.layers.78.mlp.experts.163.gate_proj.weight": "model.safetensors",
228
+ "model.layers.78.mlp.experts.163.up_proj.weight": "model.safetensors",
229
+ "model.layers.78.mlp.experts.164.down_proj.weight": "model.safetensors",
230
+ "model.layers.78.mlp.experts.164.gate_proj.weight": "model.safetensors",
231
+ "model.layers.78.mlp.experts.164.up_proj.weight": "model.safetensors",
232
+ "model.layers.78.mlp.experts.165.down_proj.weight": "model.safetensors",
233
+ "model.layers.78.mlp.experts.165.gate_proj.weight": "model.safetensors",
234
+ "model.layers.78.mlp.experts.165.up_proj.weight": "model.safetensors",
235
+ "model.layers.78.mlp.experts.166.down_proj.weight": "model.safetensors",
236
+ "model.layers.78.mlp.experts.166.gate_proj.weight": "model.safetensors",
237
+ "model.layers.78.mlp.experts.166.up_proj.weight": "model.safetensors",
238
+ "model.layers.78.mlp.experts.167.down_proj.weight": "model.safetensors",
239
+ "model.layers.78.mlp.experts.167.gate_proj.weight": "model.safetensors",
240
+ "model.layers.78.mlp.experts.167.up_proj.weight": "model.safetensors",
241
+ "model.layers.78.mlp.experts.168.down_proj.weight": "model.safetensors",
242
+ "model.layers.78.mlp.experts.168.gate_proj.weight": "model.safetensors",
243
+ "model.layers.78.mlp.experts.168.up_proj.weight": "model.safetensors",
244
+ "model.layers.78.mlp.experts.169.down_proj.weight": "model.safetensors",
245
+ "model.layers.78.mlp.experts.169.gate_proj.weight": "model.safetensors",
246
+ "model.layers.78.mlp.experts.169.up_proj.weight": "model.safetensors",
247
+ "model.layers.78.mlp.experts.17.down_proj.weight": "model.safetensors",
248
+ "model.layers.78.mlp.experts.17.gate_proj.weight": "model.safetensors",
249
+ "model.layers.78.mlp.experts.17.up_proj.weight": "model.safetensors",
250
+ "model.layers.78.mlp.experts.170.down_proj.weight": "model.safetensors",
251
+ "model.layers.78.mlp.experts.170.gate_proj.weight": "model.safetensors",
252
+ "model.layers.78.mlp.experts.170.up_proj.weight": "model.safetensors",
253
+ "model.layers.78.mlp.experts.171.down_proj.weight": "model.safetensors",
254
+ "model.layers.78.mlp.experts.171.gate_proj.weight": "model.safetensors",
255
+ "model.layers.78.mlp.experts.171.up_proj.weight": "model.safetensors",
256
+ "model.layers.78.mlp.experts.172.down_proj.weight": "model.safetensors",
257
+ "model.layers.78.mlp.experts.172.gate_proj.weight": "model.safetensors",
258
+ "model.layers.78.mlp.experts.172.up_proj.weight": "model.safetensors",
259
+ "model.layers.78.mlp.experts.173.down_proj.weight": "model.safetensors",
260
+ "model.layers.78.mlp.experts.173.gate_proj.weight": "model.safetensors",
261
+ "model.layers.78.mlp.experts.173.up_proj.weight": "model.safetensors",
262
+ "model.layers.78.mlp.experts.174.down_proj.weight": "model.safetensors",
263
+ "model.layers.78.mlp.experts.174.gate_proj.weight": "model.safetensors",
264
+ "model.layers.78.mlp.experts.174.up_proj.weight": "model.safetensors",
265
+ "model.layers.78.mlp.experts.175.down_proj.weight": "model.safetensors",
266
+ "model.layers.78.mlp.experts.175.gate_proj.weight": "model.safetensors",
267
+ "model.layers.78.mlp.experts.175.up_proj.weight": "model.safetensors",
268
+ "model.layers.78.mlp.experts.176.down_proj.weight": "model.safetensors",
269
+ "model.layers.78.mlp.experts.176.gate_proj.weight": "model.safetensors",
270
+ "model.layers.78.mlp.experts.176.up_proj.weight": "model.safetensors",
271
+ "model.layers.78.mlp.experts.177.down_proj.weight": "model.safetensors",
272
+ "model.layers.78.mlp.experts.177.gate_proj.weight": "model.safetensors",
273
+ "model.layers.78.mlp.experts.177.up_proj.weight": "model.safetensors",
274
+ "model.layers.78.mlp.experts.178.down_proj.weight": "model.safetensors",
275
+ "model.layers.78.mlp.experts.178.gate_proj.weight": "model.safetensors",
276
+ "model.layers.78.mlp.experts.178.up_proj.weight": "model.safetensors",
277
+ "model.layers.78.mlp.experts.179.down_proj.weight": "model.safetensors",
278
+ "model.layers.78.mlp.experts.179.gate_proj.weight": "model.safetensors",
279
+ "model.layers.78.mlp.experts.179.up_proj.weight": "model.safetensors",
280
+ "model.layers.78.mlp.experts.18.down_proj.weight": "model.safetensors",
281
+ "model.layers.78.mlp.experts.18.gate_proj.weight": "model.safetensors",
282
+ "model.layers.78.mlp.experts.18.up_proj.weight": "model.safetensors",
283
+ "model.layers.78.mlp.experts.180.down_proj.weight": "model.safetensors",
284
+ "model.layers.78.mlp.experts.180.gate_proj.weight": "model.safetensors",
285
+ "model.layers.78.mlp.experts.180.up_proj.weight": "model.safetensors",
286
+ "model.layers.78.mlp.experts.181.down_proj.weight": "model.safetensors",
287
+ "model.layers.78.mlp.experts.181.gate_proj.weight": "model.safetensors",
288
+ "model.layers.78.mlp.experts.181.up_proj.weight": "model.safetensors",
289
+ "model.layers.78.mlp.experts.182.down_proj.weight": "model.safetensors",
290
+ "model.layers.78.mlp.experts.182.gate_proj.weight": "model.safetensors",
291
+ "model.layers.78.mlp.experts.182.up_proj.weight": "model.safetensors",
292
+ "model.layers.78.mlp.experts.183.down_proj.weight": "model.safetensors",
293
+ "model.layers.78.mlp.experts.183.gate_proj.weight": "model.safetensors",
294
+ "model.layers.78.mlp.experts.183.up_proj.weight": "model.safetensors",
295
+ "model.layers.78.mlp.experts.184.down_proj.weight": "model.safetensors",
296
+ "model.layers.78.mlp.experts.184.gate_proj.weight": "model.safetensors",
297
+ "model.layers.78.mlp.experts.184.up_proj.weight": "model.safetensors",
298
+ "model.layers.78.mlp.experts.185.down_proj.weight": "model.safetensors",
299
+ "model.layers.78.mlp.experts.185.gate_proj.weight": "model.safetensors",
300
+ "model.layers.78.mlp.experts.185.up_proj.weight": "model.safetensors",
301
+ "model.layers.78.mlp.experts.186.down_proj.weight": "model.safetensors",
302
+ "model.layers.78.mlp.experts.186.gate_proj.weight": "model.safetensors",
303
+ "model.layers.78.mlp.experts.186.up_proj.weight": "model.safetensors",
304
+ "model.layers.78.mlp.experts.187.down_proj.weight": "model.safetensors",
305
+ "model.layers.78.mlp.experts.187.gate_proj.weight": "model.safetensors",
306
+ "model.layers.78.mlp.experts.187.up_proj.weight": "model.safetensors",
307
+ "model.layers.78.mlp.experts.188.down_proj.weight": "model.safetensors",
308
+ "model.layers.78.mlp.experts.188.gate_proj.weight": "model.safetensors",
309
+ "model.layers.78.mlp.experts.188.up_proj.weight": "model.safetensors",
310
+ "model.layers.78.mlp.experts.189.down_proj.weight": "model.safetensors",
311
+ "model.layers.78.mlp.experts.189.gate_proj.weight": "model.safetensors",
312
+ "model.layers.78.mlp.experts.189.up_proj.weight": "model.safetensors",
313
+ "model.layers.78.mlp.experts.19.down_proj.weight": "model.safetensors",
314
+ "model.layers.78.mlp.experts.19.gate_proj.weight": "model.safetensors",
315
+ "model.layers.78.mlp.experts.19.up_proj.weight": "model.safetensors",
316
+ "model.layers.78.mlp.experts.190.down_proj.weight": "model.safetensors",
317
+ "model.layers.78.mlp.experts.190.gate_proj.weight": "model.safetensors",
318
+ "model.layers.78.mlp.experts.190.up_proj.weight": "model.safetensors",
319
+ "model.layers.78.mlp.experts.191.down_proj.weight": "model.safetensors",
320
+ "model.layers.78.mlp.experts.191.gate_proj.weight": "model.safetensors",
321
+ "model.layers.78.mlp.experts.191.up_proj.weight": "model.safetensors",
322
+ "model.layers.78.mlp.experts.192.down_proj.weight": "model.safetensors",
323
+ "model.layers.78.mlp.experts.192.gate_proj.weight": "model.safetensors",
324
+ "model.layers.78.mlp.experts.192.up_proj.weight": "model.safetensors",
325
+ "model.layers.78.mlp.experts.193.down_proj.weight": "model.safetensors",
326
+ "model.layers.78.mlp.experts.193.gate_proj.weight": "model.safetensors",
327
+ "model.layers.78.mlp.experts.193.up_proj.weight": "model.safetensors",
328
+ "model.layers.78.mlp.experts.194.down_proj.weight": "model.safetensors",
329
+ "model.layers.78.mlp.experts.194.gate_proj.weight": "model.safetensors",
330
+ "model.layers.78.mlp.experts.194.up_proj.weight": "model.safetensors",
331
+ "model.layers.78.mlp.experts.195.down_proj.weight": "model.safetensors",
332
+ "model.layers.78.mlp.experts.195.gate_proj.weight": "model.safetensors",
333
+ "model.layers.78.mlp.experts.195.up_proj.weight": "model.safetensors",
334
+ "model.layers.78.mlp.experts.196.down_proj.weight": "model.safetensors",
335
+ "model.layers.78.mlp.experts.196.gate_proj.weight": "model.safetensors",
336
+ "model.layers.78.mlp.experts.196.up_proj.weight": "model.safetensors",
337
+ "model.layers.78.mlp.experts.197.down_proj.weight": "model.safetensors",
338
+ "model.layers.78.mlp.experts.197.gate_proj.weight": "model.safetensors",
339
+ "model.layers.78.mlp.experts.197.up_proj.weight": "model.safetensors",
340
+ "model.layers.78.mlp.experts.198.down_proj.weight": "model.safetensors",
341
+ "model.layers.78.mlp.experts.198.gate_proj.weight": "model.safetensors",
342
+ "model.layers.78.mlp.experts.198.up_proj.weight": "model.safetensors",
343
+ "model.layers.78.mlp.experts.199.down_proj.weight": "model.safetensors",
344
+ "model.layers.78.mlp.experts.199.gate_proj.weight": "model.safetensors",
345
+ "model.layers.78.mlp.experts.199.up_proj.weight": "model.safetensors",
346
+ "model.layers.78.mlp.experts.2.down_proj.weight": "model.safetensors",
347
+ "model.layers.78.mlp.experts.2.gate_proj.weight": "model.safetensors",
348
+ "model.layers.78.mlp.experts.2.up_proj.weight": "model.safetensors",
349
+ "model.layers.78.mlp.experts.20.down_proj.weight": "model.safetensors",
350
+ "model.layers.78.mlp.experts.20.gate_proj.weight": "model.safetensors",
351
+ "model.layers.78.mlp.experts.20.up_proj.weight": "model.safetensors",
352
+ "model.layers.78.mlp.experts.200.down_proj.weight": "model.safetensors",
353
+ "model.layers.78.mlp.experts.200.gate_proj.weight": "model.safetensors",
354
+ "model.layers.78.mlp.experts.200.up_proj.weight": "model.safetensors",
355
+ "model.layers.78.mlp.experts.201.down_proj.weight": "model.safetensors",
356
+ "model.layers.78.mlp.experts.201.gate_proj.weight": "model.safetensors",
357
+ "model.layers.78.mlp.experts.201.up_proj.weight": "model.safetensors",
358
+ "model.layers.78.mlp.experts.202.down_proj.weight": "model.safetensors",
359
+ "model.layers.78.mlp.experts.202.gate_proj.weight": "model.safetensors",
360
+ "model.layers.78.mlp.experts.202.up_proj.weight": "model.safetensors",
361
+ "model.layers.78.mlp.experts.203.down_proj.weight": "model.safetensors",
362
+ "model.layers.78.mlp.experts.203.gate_proj.weight": "model.safetensors",
363
+ "model.layers.78.mlp.experts.203.up_proj.weight": "model.safetensors",
364
+ "model.layers.78.mlp.experts.204.down_proj.weight": "model.safetensors",
365
+ "model.layers.78.mlp.experts.204.gate_proj.weight": "model.safetensors",
366
+ "model.layers.78.mlp.experts.204.up_proj.weight": "model.safetensors",
367
+ "model.layers.78.mlp.experts.205.down_proj.weight": "model.safetensors",
368
+ "model.layers.78.mlp.experts.205.gate_proj.weight": "model.safetensors",
369
+ "model.layers.78.mlp.experts.205.up_proj.weight": "model.safetensors",
370
+ "model.layers.78.mlp.experts.206.down_proj.weight": "model.safetensors",
371
+ "model.layers.78.mlp.experts.206.gate_proj.weight": "model.safetensors",
372
+ "model.layers.78.mlp.experts.206.up_proj.weight": "model.safetensors",
373
+ "model.layers.78.mlp.experts.207.down_proj.weight": "model.safetensors",
374
+ "model.layers.78.mlp.experts.207.gate_proj.weight": "model.safetensors",
375
+ "model.layers.78.mlp.experts.207.up_proj.weight": "model.safetensors",
376
+ "model.layers.78.mlp.experts.208.down_proj.weight": "model.safetensors",
377
+ "model.layers.78.mlp.experts.208.gate_proj.weight": "model.safetensors",
378
+ "model.layers.78.mlp.experts.208.up_proj.weight": "model.safetensors",
379
+ "model.layers.78.mlp.experts.209.down_proj.weight": "model.safetensors",
380
+ "model.layers.78.mlp.experts.209.gate_proj.weight": "model.safetensors",
381
+ "model.layers.78.mlp.experts.209.up_proj.weight": "model.safetensors",
382
+ "model.layers.78.mlp.experts.21.down_proj.weight": "model.safetensors",
383
+ "model.layers.78.mlp.experts.21.gate_proj.weight": "model.safetensors",
384
+ "model.layers.78.mlp.experts.21.up_proj.weight": "model.safetensors",
385
+ "model.layers.78.mlp.experts.210.down_proj.weight": "model.safetensors",
386
+ "model.layers.78.mlp.experts.210.gate_proj.weight": "model.safetensors",
387
+ "model.layers.78.mlp.experts.210.up_proj.weight": "model.safetensors",
388
+ "model.layers.78.mlp.experts.211.down_proj.weight": "model.safetensors",
389
+ "model.layers.78.mlp.experts.211.gate_proj.weight": "model.safetensors",
390
+ "model.layers.78.mlp.experts.211.up_proj.weight": "model.safetensors",
391
+ "model.layers.78.mlp.experts.212.down_proj.weight": "model.safetensors",
392
+ "model.layers.78.mlp.experts.212.gate_proj.weight": "model.safetensors",
393
+ "model.layers.78.mlp.experts.212.up_proj.weight": "model.safetensors",
394
+ "model.layers.78.mlp.experts.213.down_proj.weight": "model.safetensors",
395
+ "model.layers.78.mlp.experts.213.gate_proj.weight": "model.safetensors",
396
+ "model.layers.78.mlp.experts.213.up_proj.weight": "model.safetensors",
397
+ "model.layers.78.mlp.experts.214.down_proj.weight": "model.safetensors",
398
+ "model.layers.78.mlp.experts.214.gate_proj.weight": "model.safetensors",
399
+ "model.layers.78.mlp.experts.214.up_proj.weight": "model.safetensors",
400
+ "model.layers.78.mlp.experts.215.down_proj.weight": "model.safetensors",
401
+ "model.layers.78.mlp.experts.215.gate_proj.weight": "model.safetensors",
402
+ "model.layers.78.mlp.experts.215.up_proj.weight": "model.safetensors",
403
+ "model.layers.78.mlp.experts.216.down_proj.weight": "model.safetensors",
404
+ "model.layers.78.mlp.experts.216.gate_proj.weight": "model.safetensors",
405
+ "model.layers.78.mlp.experts.216.up_proj.weight": "model.safetensors",
406
+ "model.layers.78.mlp.experts.217.down_proj.weight": "model.safetensors",
407
+ "model.layers.78.mlp.experts.217.gate_proj.weight": "model.safetensors",
408
+ "model.layers.78.mlp.experts.217.up_proj.weight": "model.safetensors",
409
+ "model.layers.78.mlp.experts.218.down_proj.weight": "model.safetensors",
410
+ "model.layers.78.mlp.experts.218.gate_proj.weight": "model.safetensors",
411
+ "model.layers.78.mlp.experts.218.up_proj.weight": "model.safetensors",
412
+ "model.layers.78.mlp.experts.219.down_proj.weight": "model.safetensors",
413
+ "model.layers.78.mlp.experts.219.gate_proj.weight": "model.safetensors",
414
+ "model.layers.78.mlp.experts.219.up_proj.weight": "model.safetensors",
415
+ "model.layers.78.mlp.experts.22.down_proj.weight": "model.safetensors",
416
+ "model.layers.78.mlp.experts.22.gate_proj.weight": "model.safetensors",
417
+ "model.layers.78.mlp.experts.22.up_proj.weight": "model.safetensors",
418
+ "model.layers.78.mlp.experts.220.down_proj.weight": "model.safetensors",
419
+ "model.layers.78.mlp.experts.220.gate_proj.weight": "model.safetensors",
420
+ "model.layers.78.mlp.experts.220.up_proj.weight": "model.safetensors",
421
+ "model.layers.78.mlp.experts.221.down_proj.weight": "model.safetensors",
422
+ "model.layers.78.mlp.experts.221.gate_proj.weight": "model.safetensors",
423
+ "model.layers.78.mlp.experts.221.up_proj.weight": "model.safetensors",
424
+ "model.layers.78.mlp.experts.222.down_proj.weight": "model.safetensors",
425
+ "model.layers.78.mlp.experts.222.gate_proj.weight": "model.safetensors",
426
+ "model.layers.78.mlp.experts.222.up_proj.weight": "model.safetensors",
427
+ "model.layers.78.mlp.experts.223.down_proj.weight": "model.safetensors",
428
+ "model.layers.78.mlp.experts.223.gate_proj.weight": "model.safetensors",
429
+ "model.layers.78.mlp.experts.223.up_proj.weight": "model.safetensors",
430
+ "model.layers.78.mlp.experts.224.down_proj.weight": "model.safetensors",
431
+ "model.layers.78.mlp.experts.224.gate_proj.weight": "model.safetensors",
432
+ "model.layers.78.mlp.experts.224.up_proj.weight": "model.safetensors",
433
+ "model.layers.78.mlp.experts.225.down_proj.weight": "model.safetensors",
434
+ "model.layers.78.mlp.experts.225.gate_proj.weight": "model.safetensors",
435
+ "model.layers.78.mlp.experts.225.up_proj.weight": "model.safetensors",
436
+ "model.layers.78.mlp.experts.226.down_proj.weight": "model.safetensors",
437
+ "model.layers.78.mlp.experts.226.gate_proj.weight": "model.safetensors",
438
+ "model.layers.78.mlp.experts.226.up_proj.weight": "model.safetensors",
439
+ "model.layers.78.mlp.experts.227.down_proj.weight": "model.safetensors",
440
+ "model.layers.78.mlp.experts.227.gate_proj.weight": "model.safetensors",
441
+ "model.layers.78.mlp.experts.227.up_proj.weight": "model.safetensors",
442
+ "model.layers.78.mlp.experts.228.down_proj.weight": "model.safetensors",
443
+ "model.layers.78.mlp.experts.228.gate_proj.weight": "model.safetensors",
444
+ "model.layers.78.mlp.experts.228.up_proj.weight": "model.safetensors",
445
+ "model.layers.78.mlp.experts.229.down_proj.weight": "model.safetensors",
446
+ "model.layers.78.mlp.experts.229.gate_proj.weight": "model.safetensors",
447
+ "model.layers.78.mlp.experts.229.up_proj.weight": "model.safetensors",
448
+ "model.layers.78.mlp.experts.23.down_proj.weight": "model.safetensors",
449
+ "model.layers.78.mlp.experts.23.gate_proj.weight": "model.safetensors",
450
+ "model.layers.78.mlp.experts.23.up_proj.weight": "model.safetensors",
451
+ "model.layers.78.mlp.experts.230.down_proj.weight": "model.safetensors",
452
+ "model.layers.78.mlp.experts.230.gate_proj.weight": "model.safetensors",
453
+ "model.layers.78.mlp.experts.230.up_proj.weight": "model.safetensors",
454
+ "model.layers.78.mlp.experts.231.down_proj.weight": "model.safetensors",
455
+ "model.layers.78.mlp.experts.231.gate_proj.weight": "model.safetensors",
456
+ "model.layers.78.mlp.experts.231.up_proj.weight": "model.safetensors",
457
+ "model.layers.78.mlp.experts.232.down_proj.weight": "model.safetensors",
458
+ "model.layers.78.mlp.experts.232.gate_proj.weight": "model.safetensors",
459
+ "model.layers.78.mlp.experts.232.up_proj.weight": "model.safetensors",
460
+ "model.layers.78.mlp.experts.233.down_proj.weight": "model.safetensors",
461
+ "model.layers.78.mlp.experts.233.gate_proj.weight": "model.safetensors",
462
+ "model.layers.78.mlp.experts.233.up_proj.weight": "model.safetensors",
463
+ "model.layers.78.mlp.experts.234.down_proj.weight": "model.safetensors",
464
+ "model.layers.78.mlp.experts.234.gate_proj.weight": "model.safetensors",
465
+ "model.layers.78.mlp.experts.234.up_proj.weight": "model.safetensors",
466
+ "model.layers.78.mlp.experts.235.down_proj.weight": "model.safetensors",
467
+ "model.layers.78.mlp.experts.235.gate_proj.weight": "model.safetensors",
468
+ "model.layers.78.mlp.experts.235.up_proj.weight": "model.safetensors",
469
+ "model.layers.78.mlp.experts.236.down_proj.weight": "model.safetensors",
470
+ "model.layers.78.mlp.experts.236.gate_proj.weight": "model.safetensors",
471
+ "model.layers.78.mlp.experts.236.up_proj.weight": "model.safetensors",
472
+ "model.layers.78.mlp.experts.237.down_proj.weight": "model.safetensors",
473
+ "model.layers.78.mlp.experts.237.gate_proj.weight": "model.safetensors",
474
+ "model.layers.78.mlp.experts.237.up_proj.weight": "model.safetensors",
475
+ "model.layers.78.mlp.experts.238.down_proj.weight": "model.safetensors",
476
+ "model.layers.78.mlp.experts.238.gate_proj.weight": "model.safetensors",
477
+ "model.layers.78.mlp.experts.238.up_proj.weight": "model.safetensors",
478
+ "model.layers.78.mlp.experts.239.down_proj.weight": "model.safetensors",
479
+ "model.layers.78.mlp.experts.239.gate_proj.weight": "model.safetensors",
480
+ "model.layers.78.mlp.experts.239.up_proj.weight": "model.safetensors",
481
+ "model.layers.78.mlp.experts.24.down_proj.weight": "model.safetensors",
482
+ "model.layers.78.mlp.experts.24.gate_proj.weight": "model.safetensors",
483
+ "model.layers.78.mlp.experts.24.up_proj.weight": "model.safetensors",
484
+ "model.layers.78.mlp.experts.240.down_proj.weight": "model.safetensors",
485
+ "model.layers.78.mlp.experts.240.gate_proj.weight": "model.safetensors",
486
+ "model.layers.78.mlp.experts.240.up_proj.weight": "model.safetensors",
487
+ "model.layers.78.mlp.experts.241.down_proj.weight": "model.safetensors",
488
+ "model.layers.78.mlp.experts.241.gate_proj.weight": "model.safetensors",
489
+ "model.layers.78.mlp.experts.241.up_proj.weight": "model.safetensors",
490
+ "model.layers.78.mlp.experts.242.down_proj.weight": "model.safetensors",
491
+ "model.layers.78.mlp.experts.242.gate_proj.weight": "model.safetensors",
492
+ "model.layers.78.mlp.experts.242.up_proj.weight": "model.safetensors",
493
+ "model.layers.78.mlp.experts.243.down_proj.weight": "model.safetensors",
494
+ "model.layers.78.mlp.experts.243.gate_proj.weight": "model.safetensors",
495
+ "model.layers.78.mlp.experts.243.up_proj.weight": "model.safetensors",
496
+ "model.layers.78.mlp.experts.244.down_proj.weight": "model.safetensors",
497
+ "model.layers.78.mlp.experts.244.gate_proj.weight": "model.safetensors",
498
+ "model.layers.78.mlp.experts.244.up_proj.weight": "model.safetensors",
499
+ "model.layers.78.mlp.experts.245.down_proj.weight": "model.safetensors",
500
+ "model.layers.78.mlp.experts.245.gate_proj.weight": "model.safetensors",
501
+ "model.layers.78.mlp.experts.245.up_proj.weight": "model.safetensors",
502
+ "model.layers.78.mlp.experts.246.down_proj.weight": "model.safetensors",
503
+ "model.layers.78.mlp.experts.246.gate_proj.weight": "model.safetensors",
504
+ "model.layers.78.mlp.experts.246.up_proj.weight": "model.safetensors",
505
+ "model.layers.78.mlp.experts.247.down_proj.weight": "model.safetensors",
506
+ "model.layers.78.mlp.experts.247.gate_proj.weight": "model.safetensors",
507
+ "model.layers.78.mlp.experts.247.up_proj.weight": "model.safetensors",
508
+ "model.layers.78.mlp.experts.248.down_proj.weight": "model.safetensors",
509
+ "model.layers.78.mlp.experts.248.gate_proj.weight": "model.safetensors",
510
+ "model.layers.78.mlp.experts.248.up_proj.weight": "model.safetensors",
511
+ "model.layers.78.mlp.experts.249.down_proj.weight": "model.safetensors",
512
+ "model.layers.78.mlp.experts.249.gate_proj.weight": "model.safetensors",
513
+ "model.layers.78.mlp.experts.249.up_proj.weight": "model.safetensors",
514
+ "model.layers.78.mlp.experts.25.down_proj.weight": "model.safetensors",
515
+ "model.layers.78.mlp.experts.25.gate_proj.weight": "model.safetensors",
516
+ "model.layers.78.mlp.experts.25.up_proj.weight": "model.safetensors",
517
+ "model.layers.78.mlp.experts.250.down_proj.weight": "model.safetensors",
518
+ "model.layers.78.mlp.experts.250.gate_proj.weight": "model.safetensors",
519
+ "model.layers.78.mlp.experts.250.up_proj.weight": "model.safetensors",
520
+ "model.layers.78.mlp.experts.251.down_proj.weight": "model.safetensors",
521
+ "model.layers.78.mlp.experts.251.gate_proj.weight": "model.safetensors",
522
+ "model.layers.78.mlp.experts.251.up_proj.weight": "model.safetensors",
523
+ "model.layers.78.mlp.experts.252.down_proj.weight": "model.safetensors",
524
+ "model.layers.78.mlp.experts.252.gate_proj.weight": "model.safetensors",
525
+ "model.layers.78.mlp.experts.252.up_proj.weight": "model.safetensors",
526
+ "model.layers.78.mlp.experts.253.down_proj.weight": "model.safetensors",
527
+ "model.layers.78.mlp.experts.253.gate_proj.weight": "model.safetensors",
528
+ "model.layers.78.mlp.experts.253.up_proj.weight": "model.safetensors",
529
+ "model.layers.78.mlp.experts.254.down_proj.weight": "model.safetensors",
530
+ "model.layers.78.mlp.experts.254.gate_proj.weight": "model.safetensors",
531
+ "model.layers.78.mlp.experts.254.up_proj.weight": "model.safetensors",
532
+ "model.layers.78.mlp.experts.255.down_proj.weight": "model.safetensors",
533
+ "model.layers.78.mlp.experts.255.gate_proj.weight": "model.safetensors",
534
+ "model.layers.78.mlp.experts.255.up_proj.weight": "model.safetensors",
535
+ "model.layers.78.mlp.experts.26.down_proj.weight": "model.safetensors",
536
+ "model.layers.78.mlp.experts.26.gate_proj.weight": "model.safetensors",
537
+ "model.layers.78.mlp.experts.26.up_proj.weight": "model.safetensors",
538
+ "model.layers.78.mlp.experts.27.down_proj.weight": "model.safetensors",
539
+ "model.layers.78.mlp.experts.27.gate_proj.weight": "model.safetensors",
540
+ "model.layers.78.mlp.experts.27.up_proj.weight": "model.safetensors",
541
+ "model.layers.78.mlp.experts.28.down_proj.weight": "model.safetensors",
542
+ "model.layers.78.mlp.experts.28.gate_proj.weight": "model.safetensors",
543
+ "model.layers.78.mlp.experts.28.up_proj.weight": "model.safetensors",
544
+ "model.layers.78.mlp.experts.29.down_proj.weight": "model.safetensors",
545
+ "model.layers.78.mlp.experts.29.gate_proj.weight": "model.safetensors",
546
+ "model.layers.78.mlp.experts.29.up_proj.weight": "model.safetensors",
547
+ "model.layers.78.mlp.experts.3.down_proj.weight": "model.safetensors",
548
+ "model.layers.78.mlp.experts.3.gate_proj.weight": "model.safetensors",
549
+ "model.layers.78.mlp.experts.3.up_proj.weight": "model.safetensors",
550
+ "model.layers.78.mlp.experts.30.down_proj.weight": "model.safetensors",
551
+ "model.layers.78.mlp.experts.30.gate_proj.weight": "model.safetensors",
552
+ "model.layers.78.mlp.experts.30.up_proj.weight": "model.safetensors",
553
+ "model.layers.78.mlp.experts.31.down_proj.weight": "model.safetensors",
554
+ "model.layers.78.mlp.experts.31.gate_proj.weight": "model.safetensors",
555
+ "model.layers.78.mlp.experts.31.up_proj.weight": "model.safetensors",
556
+ "model.layers.78.mlp.experts.32.down_proj.weight": "model.safetensors",
557
+ "model.layers.78.mlp.experts.32.gate_proj.weight": "model.safetensors",
558
+ "model.layers.78.mlp.experts.32.up_proj.weight": "model.safetensors",
559
+ "model.layers.78.mlp.experts.33.down_proj.weight": "model.safetensors",
560
+ "model.layers.78.mlp.experts.33.gate_proj.weight": "model.safetensors",
561
+ "model.layers.78.mlp.experts.33.up_proj.weight": "model.safetensors",
562
+ "model.layers.78.mlp.experts.34.down_proj.weight": "model.safetensors",
563
+ "model.layers.78.mlp.experts.34.gate_proj.weight": "model.safetensors",
564
+ "model.layers.78.mlp.experts.34.up_proj.weight": "model.safetensors",
565
+ "model.layers.78.mlp.experts.35.down_proj.weight": "model.safetensors",
566
+ "model.layers.78.mlp.experts.35.gate_proj.weight": "model.safetensors",
567
+ "model.layers.78.mlp.experts.35.up_proj.weight": "model.safetensors",
568
+ "model.layers.78.mlp.experts.36.down_proj.weight": "model.safetensors",
569
+ "model.layers.78.mlp.experts.36.gate_proj.weight": "model.safetensors",
570
+ "model.layers.78.mlp.experts.36.up_proj.weight": "model.safetensors",
571
+ "model.layers.78.mlp.experts.37.down_proj.weight": "model.safetensors",
572
+ "model.layers.78.mlp.experts.37.gate_proj.weight": "model.safetensors",
573
+ "model.layers.78.mlp.experts.37.up_proj.weight": "model.safetensors",
574
+ "model.layers.78.mlp.experts.38.down_proj.weight": "model.safetensors",
575
+ "model.layers.78.mlp.experts.38.gate_proj.weight": "model.safetensors",
576
+ "model.layers.78.mlp.experts.38.up_proj.weight": "model.safetensors",
577
+ "model.layers.78.mlp.experts.39.down_proj.weight": "model.safetensors",
578
+ "model.layers.78.mlp.experts.39.gate_proj.weight": "model.safetensors",
579
+ "model.layers.78.mlp.experts.39.up_proj.weight": "model.safetensors",
580
+ "model.layers.78.mlp.experts.4.down_proj.weight": "model.safetensors",
581
+ "model.layers.78.mlp.experts.4.gate_proj.weight": "model.safetensors",
582
+ "model.layers.78.mlp.experts.4.up_proj.weight": "model.safetensors",
583
+ "model.layers.78.mlp.experts.40.down_proj.weight": "model.safetensors",
584
+ "model.layers.78.mlp.experts.40.gate_proj.weight": "model.safetensors",
585
+ "model.layers.78.mlp.experts.40.up_proj.weight": "model.safetensors",
586
+ "model.layers.78.mlp.experts.41.down_proj.weight": "model.safetensors",
587
+ "model.layers.78.mlp.experts.41.gate_proj.weight": "model.safetensors",
588
+ "model.layers.78.mlp.experts.41.up_proj.weight": "model.safetensors",
589
+ "model.layers.78.mlp.experts.42.down_proj.weight": "model.safetensors",
590
+ "model.layers.78.mlp.experts.42.gate_proj.weight": "model.safetensors",
591
+ "model.layers.78.mlp.experts.42.up_proj.weight": "model.safetensors",
592
+ "model.layers.78.mlp.experts.43.down_proj.weight": "model.safetensors",
593
+ "model.layers.78.mlp.experts.43.gate_proj.weight": "model.safetensors",
594
+ "model.layers.78.mlp.experts.43.up_proj.weight": "model.safetensors",
595
+ "model.layers.78.mlp.experts.44.down_proj.weight": "model.safetensors",
596
+ "model.layers.78.mlp.experts.44.gate_proj.weight": "model.safetensors",
597
+ "model.layers.78.mlp.experts.44.up_proj.weight": "model.safetensors",
598
+ "model.layers.78.mlp.experts.45.down_proj.weight": "model.safetensors",
599
+ "model.layers.78.mlp.experts.45.gate_proj.weight": "model.safetensors",
600
+ "model.layers.78.mlp.experts.45.up_proj.weight": "model.safetensors",
601
+ "model.layers.78.mlp.experts.46.down_proj.weight": "model.safetensors",
602
+ "model.layers.78.mlp.experts.46.gate_proj.weight": "model.safetensors",
603
+ "model.layers.78.mlp.experts.46.up_proj.weight": "model.safetensors",
604
+ "model.layers.78.mlp.experts.47.down_proj.weight": "model.safetensors",
605
+ "model.layers.78.mlp.experts.47.gate_proj.weight": "model.safetensors",
606
+ "model.layers.78.mlp.experts.47.up_proj.weight": "model.safetensors",
607
+ "model.layers.78.mlp.experts.48.down_proj.weight": "model.safetensors",
608
+ "model.layers.78.mlp.experts.48.gate_proj.weight": "model.safetensors",
609
+ "model.layers.78.mlp.experts.48.up_proj.weight": "model.safetensors",
610
+ "model.layers.78.mlp.experts.49.down_proj.weight": "model.safetensors",
611
+ "model.layers.78.mlp.experts.49.gate_proj.weight": "model.safetensors",
612
+ "model.layers.78.mlp.experts.49.up_proj.weight": "model.safetensors",
613
+ "model.layers.78.mlp.experts.5.down_proj.weight": "model.safetensors",
614
+ "model.layers.78.mlp.experts.5.gate_proj.weight": "model.safetensors",
615
+ "model.layers.78.mlp.experts.5.up_proj.weight": "model.safetensors",
616
+ "model.layers.78.mlp.experts.50.down_proj.weight": "model.safetensors",
617
+ "model.layers.78.mlp.experts.50.gate_proj.weight": "model.safetensors",
618
+ "model.layers.78.mlp.experts.50.up_proj.weight": "model.safetensors",
619
+ "model.layers.78.mlp.experts.51.down_proj.weight": "model.safetensors",
620
+ "model.layers.78.mlp.experts.51.gate_proj.weight": "model.safetensors",
621
+ "model.layers.78.mlp.experts.51.up_proj.weight": "model.safetensors",
622
+ "model.layers.78.mlp.experts.52.down_proj.weight": "model.safetensors",
623
+ "model.layers.78.mlp.experts.52.gate_proj.weight": "model.safetensors",
624
+ "model.layers.78.mlp.experts.52.up_proj.weight": "model.safetensors",
625
+ "model.layers.78.mlp.experts.53.down_proj.weight": "model.safetensors",
626
+ "model.layers.78.mlp.experts.53.gate_proj.weight": "model.safetensors",
627
+ "model.layers.78.mlp.experts.53.up_proj.weight": "model.safetensors",
628
+ "model.layers.78.mlp.experts.54.down_proj.weight": "model.safetensors",
629
+ "model.layers.78.mlp.experts.54.gate_proj.weight": "model.safetensors",
630
+ "model.layers.78.mlp.experts.54.up_proj.weight": "model.safetensors",
631
+ "model.layers.78.mlp.experts.55.down_proj.weight": "model.safetensors",
632
+ "model.layers.78.mlp.experts.55.gate_proj.weight": "model.safetensors",
633
+ "model.layers.78.mlp.experts.55.up_proj.weight": "model.safetensors",
634
+ "model.layers.78.mlp.experts.56.down_proj.weight": "model.safetensors",
635
+ "model.layers.78.mlp.experts.56.gate_proj.weight": "model.safetensors",
636
+ "model.layers.78.mlp.experts.56.up_proj.weight": "model.safetensors",
637
+ "model.layers.78.mlp.experts.57.down_proj.weight": "model.safetensors",
638
+ "model.layers.78.mlp.experts.57.gate_proj.weight": "model.safetensors",
639
+ "model.layers.78.mlp.experts.57.up_proj.weight": "model.safetensors",
640
+ "model.layers.78.mlp.experts.58.down_proj.weight": "model.safetensors",
641
+ "model.layers.78.mlp.experts.58.gate_proj.weight": "model.safetensors",
642
+ "model.layers.78.mlp.experts.58.up_proj.weight": "model.safetensors",
643
+ "model.layers.78.mlp.experts.59.down_proj.weight": "model.safetensors",
644
+ "model.layers.78.mlp.experts.59.gate_proj.weight": "model.safetensors",
645
+ "model.layers.78.mlp.experts.59.up_proj.weight": "model.safetensors",
646
+ "model.layers.78.mlp.experts.6.down_proj.weight": "model.safetensors",
647
+ "model.layers.78.mlp.experts.6.gate_proj.weight": "model.safetensors",
648
+ "model.layers.78.mlp.experts.6.up_proj.weight": "model.safetensors",
649
+ "model.layers.78.mlp.gate.e_score_correction_bias": "model.safetensors",
650
+ "model.layers.78.self_attn.indexer.weights_proj.weight": "model.safetensors",
651
+ "model.layers.78.shared_head.norm.weight": "model.safetensors",
652
+ "model.layers.78.mlp.experts.60.down_proj.weight": "model.safetensors",
653
+ "model.layers.78.mlp.experts.60.gate_proj.weight": "model.safetensors",
654
+ "model.layers.78.mlp.experts.60.up_proj.weight": "model.safetensors",
655
+ "model.layers.78.mlp.experts.61.down_proj.weight": "model.safetensors",
656
+ "model.layers.78.mlp.experts.61.gate_proj.weight": "model.safetensors",
657
+ "model.layers.78.mlp.experts.61.up_proj.weight": "model.safetensors",
658
+ "model.layers.78.mlp.experts.62.down_proj.weight": "model.safetensors",
659
+ "model.layers.78.mlp.experts.62.gate_proj.weight": "model.safetensors",
660
+ "model.layers.78.mlp.experts.62.up_proj.weight": "model.safetensors",
661
+ "model.layers.78.mlp.experts.63.down_proj.weight": "model.safetensors",
662
+ "model.layers.78.mlp.experts.63.gate_proj.weight": "model.safetensors",
663
+ "model.layers.78.mlp.experts.63.up_proj.weight": "model.safetensors",
664
+ "model.layers.78.mlp.experts.64.down_proj.weight": "model.safetensors",
665
+ "model.layers.78.mlp.experts.64.gate_proj.weight": "model.safetensors",
666
+ "model.layers.78.mlp.experts.64.up_proj.weight": "model.safetensors",
667
+ "model.layers.78.mlp.experts.65.down_proj.weight": "model.safetensors",
668
+ "model.layers.78.mlp.experts.65.gate_proj.weight": "model.safetensors",
669
+ "model.layers.78.mlp.experts.65.up_proj.weight": "model.safetensors",
670
+ "model.layers.78.mlp.experts.66.down_proj.weight": "model.safetensors",
671
+ "model.layers.78.mlp.experts.66.gate_proj.weight": "model.safetensors",
672
+ "model.layers.78.mlp.experts.66.up_proj.weight": "model.safetensors",
673
+ "model.layers.78.mlp.experts.67.down_proj.weight": "model.safetensors",
674
+ "model.layers.78.mlp.experts.67.gate_proj.weight": "model.safetensors",
675
+ "model.layers.78.mlp.experts.67.up_proj.weight": "model.safetensors",
676
+ "model.layers.78.mlp.experts.68.down_proj.weight": "model.safetensors",
677
+ "model.layers.78.mlp.experts.68.gate_proj.weight": "model.safetensors",
678
+ "model.layers.78.mlp.experts.68.up_proj.weight": "model.safetensors",
679
+ "model.layers.78.mlp.experts.69.down_proj.weight": "model.safetensors",
680
+ "model.layers.78.mlp.experts.69.gate_proj.weight": "model.safetensors",
681
+ "model.layers.78.mlp.experts.69.up_proj.weight": "model.safetensors",
682
+ "model.layers.78.mlp.experts.7.down_proj.weight": "model.safetensors",
683
+ "model.layers.78.mlp.experts.7.gate_proj.weight": "model.safetensors",
684
+ "model.layers.78.mlp.experts.7.up_proj.weight": "model.safetensors",
685
+ "model.layers.78.mlp.experts.70.down_proj.weight": "model.safetensors",
686
+ "model.layers.78.mlp.experts.70.gate_proj.weight": "model.safetensors",
687
+ "model.layers.78.mlp.experts.70.up_proj.weight": "model.safetensors",
688
+ "model.layers.78.mlp.experts.71.down_proj.weight": "model.safetensors",
689
+ "model.layers.78.mlp.experts.71.gate_proj.weight": "model.safetensors",
690
+ "model.layers.78.mlp.experts.71.up_proj.weight": "model.safetensors",
691
+ "model.layers.78.mlp.experts.72.down_proj.weight": "model.safetensors",
692
+ "model.layers.78.mlp.experts.72.gate_proj.weight": "model.safetensors",
693
+ "model.layers.78.mlp.experts.72.up_proj.weight": "model.safetensors",
694
+ "model.layers.78.mlp.experts.73.down_proj.weight": "model.safetensors",
695
+ "model.layers.78.mlp.experts.73.gate_proj.weight": "model.safetensors",
696
+ "model.layers.78.mlp.experts.73.up_proj.weight": "model.safetensors",
697
+ "model.layers.78.mlp.experts.74.down_proj.weight": "model.safetensors",
698
+ "model.layers.78.mlp.experts.74.gate_proj.weight": "model.safetensors",
699
+ "model.layers.78.mlp.experts.74.up_proj.weight": "model.safetensors",
700
+ "model.layers.78.mlp.experts.75.down_proj.weight": "model.safetensors",
701
+ "model.layers.78.mlp.experts.75.gate_proj.weight": "model.safetensors",
702
+ "model.layers.78.mlp.experts.75.up_proj.weight": "model.safetensors",
703
+ "model.layers.78.mlp.experts.76.down_proj.weight": "model.safetensors",
704
+ "model.layers.78.mlp.experts.76.gate_proj.weight": "model.safetensors",
705
+ "model.layers.78.mlp.experts.76.up_proj.weight": "model.safetensors",
706
+ "model.layers.78.mlp.experts.77.down_proj.weight": "model.safetensors",
707
+ "model.layers.78.mlp.experts.77.gate_proj.weight": "model.safetensors",
708
+ "model.layers.78.mlp.experts.77.up_proj.weight": "model.safetensors",
709
+ "model.layers.78.mlp.experts.78.down_proj.weight": "model.safetensors",
710
+ "model.layers.78.mlp.experts.78.gate_proj.weight": "model.safetensors",
711
+ "model.layers.78.mlp.experts.78.up_proj.weight": "model.safetensors",
712
+ "model.layers.78.mlp.experts.79.down_proj.weight": "model.safetensors",
713
+ "model.layers.78.mlp.experts.79.gate_proj.weight": "model.safetensors",
714
+ "model.layers.78.mlp.experts.79.up_proj.weight": "model.safetensors",
715
+ "model.layers.78.mlp.experts.8.down_proj.weight": "model.safetensors",
716
+ "model.layers.78.mlp.experts.8.gate_proj.weight": "model.safetensors",
717
+ "model.layers.78.mlp.experts.8.up_proj.weight": "model.safetensors",
718
+ "model.layers.78.mlp.experts.80.down_proj.weight": "model.safetensors",
719
+ "model.layers.78.mlp.experts.80.gate_proj.weight": "model.safetensors",
720
+ "model.layers.78.mlp.experts.80.up_proj.weight": "model.safetensors",
721
+ "model.layers.78.mlp.experts.81.down_proj.weight": "model.safetensors",
722
+ "model.layers.78.mlp.experts.81.gate_proj.weight": "model.safetensors",
723
+ "model.layers.78.mlp.experts.81.up_proj.weight": "model.safetensors",
724
+ "model.layers.78.mlp.experts.82.down_proj.weight": "model.safetensors",
725
+ "model.layers.78.mlp.experts.82.gate_proj.weight": "model.safetensors",
726
+ "model.layers.78.mlp.experts.82.up_proj.weight": "model.safetensors",
727
+ "model.layers.78.mlp.experts.83.down_proj.weight": "model.safetensors",
728
+ "model.layers.78.mlp.experts.83.gate_proj.weight": "model.safetensors",
729
+ "model.layers.78.mlp.experts.83.up_proj.weight": "model.safetensors",
730
+ "model.layers.78.mlp.experts.84.down_proj.weight": "model.safetensors",
731
+ "model.layers.78.mlp.experts.84.gate_proj.weight": "model.safetensors",
732
+ "model.layers.78.mlp.experts.84.up_proj.weight": "model.safetensors",
733
+ "model.layers.78.mlp.experts.85.down_proj.weight": "model.safetensors",
734
+ "model.layers.78.mlp.experts.85.gate_proj.weight": "model.safetensors",
735
+ "model.layers.78.mlp.experts.85.up_proj.weight": "model.safetensors",
736
+ "model.layers.78.mlp.experts.86.down_proj.weight": "model.safetensors",
737
+ "model.layers.78.mlp.experts.86.gate_proj.weight": "model.safetensors",
738
+ "model.layers.78.mlp.experts.86.up_proj.weight": "model.safetensors",
739
+ "model.layers.78.mlp.experts.87.down_proj.weight": "model.safetensors",
740
+ "model.layers.78.mlp.experts.87.gate_proj.weight": "model.safetensors",
741
+ "model.layers.78.mlp.experts.87.up_proj.weight": "model.safetensors",
742
+ "model.layers.78.mlp.experts.88.down_proj.weight": "model.safetensors",
743
+ "model.layers.78.mlp.experts.88.gate_proj.weight": "model.safetensors",
744
+ "model.layers.78.mlp.experts.88.up_proj.weight": "model.safetensors",
745
+ "model.layers.78.mlp.experts.89.down_proj.weight": "model.safetensors",
746
+ "model.layers.78.mlp.experts.89.gate_proj.weight": "model.safetensors",
747
+ "model.layers.78.mlp.experts.89.up_proj.weight": "model.safetensors",
748
+ "model.layers.78.mlp.experts.9.down_proj.weight": "model.safetensors",
749
+ "model.layers.78.mlp.experts.9.gate_proj.weight": "model.safetensors",
750
+ "model.layers.78.mlp.experts.9.up_proj.weight": "model.safetensors",
751
+ "model.layers.78.mlp.experts.90.down_proj.weight": "model.safetensors",
752
+ "model.layers.78.mlp.experts.90.gate_proj.weight": "model.safetensors",
753
+ "model.layers.78.mlp.experts.90.up_proj.weight": "model.safetensors",
754
+ "model.layers.78.mlp.experts.91.down_proj.weight": "model.safetensors",
755
+ "model.layers.78.mlp.experts.91.gate_proj.weight": "model.safetensors",
756
+ "model.layers.78.mlp.experts.91.up_proj.weight": "model.safetensors",
757
+ "model.layers.78.mlp.experts.92.down_proj.weight": "model.safetensors",
758
+ "model.layers.78.mlp.experts.92.gate_proj.weight": "model.safetensors",
759
+ "model.layers.78.mlp.experts.92.up_proj.weight": "model.safetensors",
760
+ "model.layers.78.mlp.experts.93.down_proj.weight": "model.safetensors",
761
+ "model.layers.78.mlp.experts.93.gate_proj.weight": "model.safetensors",
762
+ "model.layers.78.mlp.experts.93.up_proj.weight": "model.safetensors",
763
+ "model.layers.78.mlp.experts.94.down_proj.weight": "model.safetensors",
764
+ "model.layers.78.mlp.experts.94.gate_proj.weight": "model.safetensors",
765
+ "model.layers.78.mlp.experts.94.up_proj.weight": "model.safetensors",
766
+ "model.layers.78.mlp.experts.95.down_proj.weight": "model.safetensors",
767
+ "model.layers.78.mlp.experts.95.gate_proj.weight": "model.safetensors",
768
+ "model.layers.78.mlp.experts.95.up_proj.weight": "model.safetensors",
769
+ "model.layers.78.mlp.experts.96.down_proj.weight": "model.safetensors",
770
+ "model.layers.78.mlp.experts.96.gate_proj.weight": "model.safetensors",
771
+ "model.layers.78.mlp.experts.96.up_proj.weight": "model.safetensors",
772
+ "model.layers.78.mlp.experts.97.down_proj.weight": "model.safetensors",
773
+ "model.layers.78.mlp.experts.97.gate_proj.weight": "model.safetensors",
774
+ "model.layers.78.mlp.experts.97.up_proj.weight": "model.safetensors",
775
+ "model.layers.78.mlp.experts.98.down_proj.weight": "model.safetensors",
776
+ "model.layers.78.mlp.experts.98.gate_proj.weight": "model.safetensors",
777
+ "model.layers.78.mlp.experts.98.up_proj.weight": "model.safetensors",
778
+ "model.layers.78.mlp.experts.99.down_proj.weight": "model.safetensors",
779
+ "model.layers.78.mlp.experts.99.gate_proj.weight": "model.safetensors",
780
+ "model.layers.78.mlp.experts.99.up_proj.weight": "model.safetensors",
781
+ "model.layers.78.mlp.gate.weight": "model.safetensors",
782
+ "model.layers.78.mlp.shared_experts.down_proj.weight": "model.safetensors",
783
+ "model.layers.78.mlp.shared_experts.gate_proj.weight": "model.safetensors",
784
+ "model.layers.78.mlp.shared_experts.up_proj.weight": "model.safetensors",
785
+ "model.layers.78.post_attention_layernorm.weight": "model.safetensors",
786
+ "model.layers.78.self_attn.indexer.k_norm.bias": "model.safetensors",
787
+ "model.layers.78.self_attn.indexer.k_norm.weight": "model.safetensors",
788
+ "model.layers.78.self_attn.indexer.wk.weight": "model.safetensors",
789
+ "model.layers.78.self_attn.indexer.wq_b.weight": "model.safetensors",
790
+ "model.layers.78.self_attn.kv_a_layernorm.weight": "model.safetensors",
791
+ "model.layers.78.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
792
+ "model.layers.78.self_attn.kv_b_proj.weight": "model.safetensors",
793
+ "model.layers.78.self_attn.o_proj.weight": "model.safetensors",
794
+ "model.layers.78.self_attn.q_a_layernorm.weight": "model.safetensors",
795
+ "model.layers.78.self_attn.q_a_proj.weight": "model.safetensors",
796
+ "model.layers.78.self_attn.q_b_proj.weight": "model.safetensors"
797
+ }
798
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:19e773648cb4e65de8660ea6365e10acca112d42a854923df93db4a6f333a82d
3
+ size 20217442
tokenizer_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "clean_up_tokenization_spaces": false,
4
+ "do_lower_case": false,
5
+ "eos_token": "<|endoftext|>",
6
+ "extra_special_tokens": [
7
+ "<|endoftext|>",
8
+ "[MASK]",
9
+ "[gMASK]",
10
+ "[sMASK]",
11
+ "<sop>",
12
+ "<eop>",
13
+ "<|system|>",
14
+ "<|user|>",
15
+ "<|assistant|>",
16
+ "<|observation|>",
17
+ "<|begin_of_image|>",
18
+ "<|end_of_image|>",
19
+ "<|begin_of_video|>",
20
+ "<|end_of_video|>",
21
+ "<|begin_of_audio|>",
22
+ "<|end_of_audio|>",
23
+ "<|begin_of_transcription|>",
24
+ "<|end_of_transcription|>"
25
+ ],
26
+ "is_local": true,
27
+ "model_max_length": 1048576,
28
+ "model_specific_special_tokens": {},
29
+ "pad_token": "<|endoftext|>",
30
+ "padding_side": "left",
31
+ "remove_space": false,
32
+ "tokenizer_class": "TokenizersBackend"
33
+ }