Hugging Face's logo

ycchen
/
proof-pilot-deploy-bundle

Safetensors
proof-pilot
olmo3
dflash
speculative-decoding
on-policy-distillation
gptq
quantization
Model card Files
xet
 Community
ycchen commited on
Commit
5f48897
·
verified ·
1 Parent(s): ef8af61

Add files using upload-large-folder tool

Browse files
Files changed (12) hide show
  1. opd-32b-v33-s200-gptq-w4a16/chat_template.jinja +23 -0
  2. opd-32b-v33-s200-gptq-w4a16/config.json +221 -0
  3. opd-32b-v33-s200-gptq-w4a16/generation_config.json +11 -0
  4. opd-32b-v33-s200-gptq-w4a16/model-00001-of-00004.safetensors +3 -0
  5. opd-32b-v33-s200-gptq-w4a16/model-00002-of-00004.safetensors +3 -0
  6. opd-32b-v33-s200-gptq-w4a16/model-00003-of-00004.safetensors +3 -0
  7. opd-32b-v33-s200-gptq-w4a16/model-00004-of-00004.safetensors +3 -0
  8. opd-32b-v33-s200-gptq-w4a16/model.safetensors.index.json +0 -0
  9. opd-32b-v33-s200-gptq-w4a16/recipe.yaml +24 -0
  10. opd-32b-v33-s200-gptq-w4a16/special_tokens_map.json +5 -0
  11. opd-32b-v33-s200-gptq-w4a16/tokenizer.json +0 -0
  12. opd-32b-v33-s200-gptq-w4a16/tokenizer_config.json +14 -0
opd-32b-v33-s200-gptq-w4a16/chat_template.jinja ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {{- bos_token -}}
2
+ {%- if messages and messages[0]['role'] == 'system' -%}
3
+ {{- messages[0]['content'] -}}
4
+ {%- set loop_messages = messages[1:] -%}
5
+ {%- else -%}
6
+ {%- set loop_messages = messages -%}
7
+ {%- endif -%}
8
+ {%- for message in loop_messages -%}
9
+ {%- if message['role'] == 'user' -%}
10
+ {{- '<｜User｜>' + message['content'] -}}
11
+ {%- elif message['role'] == 'assistant' -%}
12
+ {{- '<｜Assistant｜>' -}}
13
+ {%- if message.get('reasoning_content') -%}
14
+ {{- '<think>' + message['reasoning_content'] + '</think>' -}}
15
+ {%- else -%}
16
+ {{- '</think>' -}}
17
+ {%- endif -%}
18
+ {{- message['content'] + eos_token -}}
19
+ {%- endif -%}
20
+ {%- endfor -%}
21
+ {%- if add_generation_prompt -%}
22
+ {{- '<｜Assistant｜><think>' -}}
23
+ {%- endif -%}
opd-32b-v33-s200-gptq-w4a16/config.json ADDED
@@ -0,0 +1,221 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Olmo3SinkForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 0,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 1,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 5120,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 27648,
14
+ "layer_types": [
15
+ "sliding_attention",
16
+ "sliding_attention",
17
+ "sliding_attention",
18
+ "full_attention",
19
+ "sliding_attention",
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "full_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "sliding_attention",
26
+ "full_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "full_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "full_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "full_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "full_attention",
47
+ "sliding_attention",
48
+ "sliding_attention",
49
+ "sliding_attention",
50
+ "full_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "full_attention",
55
+ "sliding_attention",
56
+ "sliding_attention",
57
+ "sliding_attention",
58
+ "full_attention",
59
+ "sliding_attention",
60
+ "sliding_attention",
61
+ "sliding_attention",
62
+ "full_attention",
63
+ "sliding_attention",
64
+ "sliding_attention",
65
+ "sliding_attention",
66
+ "full_attention",
67
+ "sliding_attention",
68
+ "sliding_attention",
69
+ "sliding_attention",
70
+ "full_attention",
71
+ "sliding_attention",
72
+ "sliding_attention",
73
+ "sliding_attention",
74
+ "full_attention",
75
+ "sliding_attention",
76
+ "sliding_attention",
77
+ "sliding_attention",
78
+ "full_attention"
79
+ ],
80
+ "max_position_embeddings": 262144,
81
+ "model_type": "olmo3",
82
+ "num_attention_heads": 40,
83
+ "num_hidden_layers": 64,
84
+ "num_key_value_heads": 8,
85
+ "pad_token_id": 2,
86
+ "reuse_packing_metadata": true,
87
+ "rms_norm_eps": 1e-06,
88
+ "sink_init_value": 0.0,
89
+ "sliding_window": 4096,
90
+ "tie_word_embeddings": false,
91
+ "transformers_version": "5.9.0",
92
+ "use_cache": true,
93
+ "vocab_size": 129280,
94
+ "torch_dtype": "bfloat16",
95
+ "rope_theta": 500000,
96
+ "rope_scaling": {
97
+ "attention_factor": 1.3465735902799727,
98
+ "beta_fast": 32.0,
99
+ "beta_slow": 1.0,
100
+ "factor": 32.0,
101
+ "original_max_position_embeddings": 8192,
102
+ "rope_type": "yarn"
103
+ },
104
+ "is_hybrid_swa": true,
105
+ "hybrid_layer_pattern": [
106
+ 1,
107
+ 1,
108
+ 1,
109
+ 0,
110
+ 1,
111
+ 1,
112
+ 1,
113
+ 0,
114
+ 1,
115
+ 1,
116
+ 1,
117
+ 0,
118
+ 1,
119
+ 1,
120
+ 1,
121
+ 0,
122
+ 1,
123
+ 1,
124
+ 1,
125
+ 0,
126
+ 1,
127
+ 1,
128
+ 1,
129
+ 0,
130
+ 1,
131
+ 1,
132
+ 1,
133
+ 0,
134
+ 1,
135
+ 1,
136
+ 1,
137
+ 0,
138
+ 1,
139
+ 1,
140
+ 1,
141
+ 0,
142
+ 1,
143
+ 1,
144
+ 1,
145
+ 0,
146
+ 1,
147
+ 1,
148
+ 1,
149
+ 0,
150
+ 1,
151
+ 1,
152
+ 1,
153
+ 0,
154
+ 1,
155
+ 1,
156
+ 1,
157
+ 0,
158
+ 1,
159
+ 1,
160
+ 1,
161
+ 0,
162
+ 1,
163
+ 1,
164
+ 1,
165
+ 0,
166
+ 1,
167
+ 1,
168
+ 1,
169
+ 0
170
+ ],
171
+ "quantization_config": {
172
+ "config_groups": {
173
+ "group_0": {
174
+ "format": "pack-quantized",
175
+ "input_activations": null,
176
+ "output_activations": null,
177
+ "targets": [
178
+ "Linear"
179
+ ],
180
+ "weights": {
181
+ "actorder": "static",
182
+ "block_structure": null,
183
+ "dynamic": false,
184
+ "group_size": 128,
185
+ "num_bits": 4,
186
+ "observer": "memoryless_minmax",
187
+ "observer_kwargs": {},
188
+ "scale_dtype": null,
189
+ "strategy": "group",
190
+ "symmetric": true,
191
+ "type": "int",
192
+ "zp_dtype": null
193
+ }
194
+ }
195
+ },
196
+ "format": "pack-quantized",
197
+ "global_compression_ratio": null,
198
+ "ignore": [
199
+ "lm_head"
200
+ ],
201
+ "kv_cache_scheme": {
202
+ "actorder": null,
203
+ "block_structure": null,
204
+ "dynamic": false,
205
+ "group_size": null,
206
+ "num_bits": 8,
207
+ "observer": "memoryless_minmax",
208
+ "observer_kwargs": {},
209
+ "scale_dtype": null,
210
+ "strategy": "tensor",
211
+ "symmetric": true,
212
+ "type": "float",
213
+ "zp_dtype": null
214
+ },
215
+ "quant_method": "compressed-tensors",
216
+ "quantization_status": "compressed",
217
+ "sparsity_config": {},
218
+ "transform_config": {},
219
+ "version": "0.16.0"
220
+ }
221
+ }
opd-32b-v33-s200-gptq-w4a16/generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "do_sample": true,
5
+ "eos_token_id": 1,
6
+ "max_new_tokens": 32768,
7
+ "pad_token_id": 2,
8
+ "temperature": 0.6,
9
+ "top_p": 0.95,
10
+ "transformers_version": "5.9.0"
11
+ }
opd-32b-v33-s200-gptq-w4a16/model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce518b31c98222d1c67d317f300b5d7b66125d68a6a4f0163812f1aa267a1b96
3
+ size 4983683172
opd-32b-v33-s200-gptq-w4a16/model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd9c4ce7fb1fc886c66fac897baf7f64fb9470a5ba946b642b26f4eb126c4bec
3
+ size 4955981384
opd-32b-v33-s200-gptq-w4a16/model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac167b9437b58433c513a90233ecf86c45b173bd825051407a19b23902f79144
3
+ size 4999223774
opd-32b-v33-s200-gptq-w4a16/model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0be74e46f2932d35ad217de565f1a395fff7abe501d408e17e21a46f558afcc5
3
+ size 3801488254
opd-32b-v33-s200-gptq-w4a16/model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
opd-32b-v33-s200-gptq-w4a16/recipe.yaml ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ default_stage:
2
+ default_modifiers:
3
+ GPTQModifier:
4
+ targets: [Linear]
5
+ ignore: [lm_head]
6
+ scheme: W4A16
7
+ kv_cache_scheme:
8
+ num_bits: 8
9
+ type: float
10
+ symmetric: true
11
+ group_size: null
12
+ strategy: tensor
13
+ block_structure: null
14
+ dynamic: false
15
+ actorder: null
16
+ scale_dtype: null
17
+ zp_dtype: null
18
+ observer: memoryless_minmax
19
+ observer_kwargs: {}
20
+ bypass_divisibility_checks: false
21
+ block_size: 128
22
+ dampening_frac: 0.01
23
+ actorder: static
24
+ offload_hessians: false
opd-32b-v33-s200-gptq-w4a16/special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<｜begin▁of▁sentence｜>",
3
+ "eos_token": "<｜end▁of▁sentence｜>",
4
+ "pad_token": "<｜▁pad▁｜>"
5
+ }
opd-32b-v33-s200-gptq-w4a16/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
opd-32b-v33-s200-gptq-w4a16/tokenizer_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "bos_token": "<｜begin▁of▁sentence｜>",
4
+ "clean_up_tokenization_spaces": false,
5
+ "eos_token": "<｜end▁of▁sentence｜>",
6
+ "is_local": true,
7
+ "legacy": true,
8
+ "local_files_only": false,
9
+ "model_max_length": 1048576,
10
+ "pad_token": "<｜▁pad▁｜>",
11
+ "sp_model_kwargs": {},
12
+ "tokenizer_class": "TokenizersBackend",
13
+ "unk_token": null
14
+ }