Hugging Face's logo

EdgeCompress01
/
Qwen3-4B-SparseGPT-GPTQ-4bit

Safetensors
Model card Files
xet
 Community
Abdohaaland commited on
Commit
316b026
·
verified ·
1 Parent(s): f5a0f3a

Upload folder using huggingface_hub

Browse files
Files changed (9) hide show
  1. .gitattributes +1 -0
  2. Models/70/chat_template.jinja +89 -0
  3. Models/70/config.json +115 -0
  4. Models/70/generation_config.json +12 -0
  5. Models/70/model.safetensors +3 -0
  6. Models/70/quant_log.csv +253 -0
  7. Models/70/quantize_config.json +44 -0
  8. Models/70/tokenizer.json +3 -0
  9. Models/70/tokenizer_config.json +29 -0
.gitattributes CHANGED
@@ -35,3 +35,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
  Models/25/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
  Models/50/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
  Models/25/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
  Models/50/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
+ Models/70/tokenizer.json filter=lfs diff=lfs merge=lfs -text
Models/70/chat_template.jinja ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0].role == 'system' %}
4
+ {{- messages[0].content + '\n\n' }}
5
+ {%- endif %}
6
+ {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
7
+ {%- for tool in tools %}
8
+ {{- "\n" }}
9
+ {{- tool | tojson }}
10
+ {%- endfor %}
11
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
12
+ {%- else %}
13
+ {%- if messages[0].role == 'system' %}
14
+ {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
15
+ {%- endif %}
16
+ {%- endif %}
17
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
18
+ {%- for message in messages[::-1] %}
19
+ {%- set index = (messages|length - 1) - loop.index0 %}
20
+ {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
21
+ {%- set ns.multi_step_tool = false %}
22
+ {%- set ns.last_query_index = index %}
23
+ {%- endif %}
24
+ {%- endfor %}
25
+ {%- for message in messages %}
26
+ {%- if message.content is string %}
27
+ {%- set content = message.content %}
28
+ {%- else %}
29
+ {%- set content = '' %}
30
+ {%- endif %}
31
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
32
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
33
+ {%- elif message.role == "assistant" %}
34
+ {%- set reasoning_content = '' %}
35
+ {%- if message.reasoning_content is string %}
36
+ {%- set reasoning_content = message.reasoning_content %}
37
+ {%- else %}
38
+ {%- if '</think>' in content %}
39
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
40
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
41
+ {%- endif %}
42
+ {%- endif %}
43
+ {%- if loop.index0 > ns.last_query_index %}
44
+ {%- if loop.last or (not loop.last and reasoning_content) %}
45
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
46
+ {%- else %}
47
+ {{- '<|im_start|>' + message.role + '\n' + content }}
48
+ {%- endif %}
49
+ {%- else %}
50
+ {{- '<|im_start|>' + message.role + '\n' + content }}
51
+ {%- endif %}
52
+ {%- if message.tool_calls %}
53
+ {%- for tool_call in message.tool_calls %}
54
+ {%- if (loop.first and content) or (not loop.first) %}
55
+ {{- '\n' }}
56
+ {%- endif %}
57
+ {%- if tool_call.function %}
58
+ {%- set tool_call = tool_call.function %}
59
+ {%- endif %}
60
+ {{- '<tool_call>\n{"name": "' }}
61
+ {{- tool_call.name }}
62
+ {{- '", "arguments": ' }}
63
+ {%- if tool_call.arguments is string %}
64
+ {{- tool_call.arguments }}
65
+ {%- else %}
66
+ {{- tool_call.arguments | tojson }}
67
+ {%- endif %}
68
+ {{- '}\n</tool_call>' }}
69
+ {%- endfor %}
70
+ {%- endif %}
71
+ {{- '<|im_end|>\n' }}
72
+ {%- elif message.role == "tool" %}
73
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
74
+ {{- '<|im_start|>user' }}
75
+ {%- endif %}
76
+ {{- '\n<tool_response>\n' }}
77
+ {{- content }}
78
+ {{- '\n</tool_response>' }}
79
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
80
+ {{- '<|im_end|>\n' }}
81
+ {%- endif %}
82
+ {%- endif %}
83
+ {%- endfor %}
84
+ {%- if add_generation_prompt %}
85
+ {{- '<|im_start|>assistant\n' }}
86
+ {%- if enable_thinking is defined and enable_thinking is false %}
87
+ {{- '<think>\n\n</think>\n\n' }}
88
+ {%- endif %}
89
+ {%- endif %}
Models/70/config.json ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151645,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2560,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 9728,
15
+ "layer_types": [
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention"
52
+ ],
53
+ "max_position_embeddings": 40960,
54
+ "max_window_layers": 36,
55
+ "model_type": "qwen3",
56
+ "num_attention_heads": 32,
57
+ "num_hidden_layers": 36,
58
+ "num_key_value_heads": 8,
59
+ "pad_token_id": 151645,
60
+ "quantization_config": {
61
+ "bits": 4,
62
+ "checkpoint_format": "gptq",
63
+ "desc_act": false,
64
+ "format": "gptq",
65
+ "group_size": 128,
66
+ "lm_head": false,
67
+ "meta": {
68
+ "act_group_aware": true,
69
+ "auto_forward_data_parallel": true,
70
+ "damp_auto_increment": 0.01,
71
+ "damp_percent": 0.05,
72
+ "fallback": {
73
+ "smooth": null,
74
+ "strategy": "rtn",
75
+ "threshold": "0.5%"
76
+ },
77
+ "foem": null,
78
+ "gc_mode": "interval",
79
+ "gptaq": null,
80
+ "hessian": {
81
+ "chunk_bytes": null,
82
+ "chunk_size": null,
83
+ "staging_dtype": "float32"
84
+ },
85
+ "mock_quantization": false,
86
+ "mse": 0.0,
87
+ "offload_to_disk": true,
88
+ "offload_to_disk_path": "./gptqmodel_offload/hkydshso-ldwypfzp/",
89
+ "pack_impl": "cpu",
90
+ "quantizer": [
91
+ "gptqmodel:6.0.0"
92
+ ],
93
+ "static_groups": false,
94
+ "true_sequential": true,
95
+ "uri": "https://github.com/modelcloud/gptqmodel",
96
+ "vram_strategy": "exclusive",
97
+ "wait_for_submodule_finalizers": false
98
+ },
99
+ "method": "gptq",
100
+ "pack_dtype": "int32",
101
+ "quant_method": "gptq",
102
+ "sym": true
103
+ },
104
+ "rms_norm_eps": 1e-06,
105
+ "rope_parameters": {
106
+ "rope_theta": 1000000,
107
+ "rope_type": "default"
108
+ },
109
+ "sliding_window": null,
110
+ "tie_word_embeddings": true,
111
+ "transformers_version": "5.5.0",
112
+ "use_cache": true,
113
+ "use_sliding_window": false,
114
+ "vocab_size": 151936
115
+ }
Models/70/generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "temperature": 0.6,
9
+ "top_k": 20,
10
+ "top_p": 0.95,
11
+ "transformers_version": "5.5.0"
12
+ }
Models/70/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0fd56e1d4241c7d01c4783a79f66328b6f7d401c9393d84c1e7e5b21d74710f2
3
+ size 2669888992
Models/70/quant_log.csv ADDED
@@ -0,0 +1,253 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ layer,module,loss,samples,damp,time
2
+ 0,self_attn.v_proj,0.0000000281,0.05000,4.670
3
+ 0,self_attn.k_proj,0.0000000335,0.05000,4.718
4
+ 0,self_attn.q_proj,0.0000001225,0.05000,4.760
5
+ 0,self_attn.o_proj,0.0000000638,0.05000,1.479
6
+ 0,mlp.gate_proj,0.0000083850,0.05000,1.913
7
+ 0,mlp.up_proj,0.0000072816,0.05000,1.916
8
+ 0,mlp.down_proj,0.0000004270,0.05000,3.658
9
+ 1,self_attn.v_proj,0.0000000562,0.05000,4.730
10
+ 1,self_attn.q_proj,0.0000002102,0.05000,4.830
11
+ 1,self_attn.k_proj,0.0000000580,0.05000,4.839
12
+ 1,self_attn.o_proj,0.0000000563,0.05000,1.447
13
+ 1,mlp.gate_proj,0.0002669175,0.05000,2.458
14
+ 1,mlp.up_proj,0.0001192833,0.05000,2.516
15
+ 1,mlp.down_proj,0.0000007279,0.05000,3.921
16
+ 2,self_attn.v_proj,0.0000001416,0.05000,4.924
17
+ 2,self_attn.q_proj,0.0000005442,0.05000,4.999
18
+ 2,self_attn.k_proj,0.0000001599,0.05000,5.030
19
+ 2,self_attn.o_proj,0.0000000774,0.05000,1.461
20
+ 2,mlp.gate_proj,0.0004706195,0.05000,2.073
21
+ 2,mlp.up_proj,0.0003679584,0.05000,2.095
22
+ 2,mlp.down_proj,0.0000005296,0.05000,3.701
23
+ 3,self_attn.v_proj,0.0000002583,0.05000,4.187
24
+ 3,self_attn.k_proj,0.0000002787,0.05000,4.232
25
+ 3,self_attn.q_proj,0.0000010549,0.05000,4.253
26
+ 3,self_attn.o_proj,0.0000001789,0.05000,1.442
27
+ 3,mlp.up_proj,0.0001859692,0.05000,1.984
28
+ 3,mlp.gate_proj,0.0003880208,0.05000,2.003
29
+ 3,mlp.down_proj,0.0000008290,0.05000,3.716
30
+ 4,self_attn.v_proj,0.0000004453,0.05000,4.170
31
+ 4,self_attn.q_proj,0.0000017503,0.05000,4.240
32
+ 4,self_attn.k_proj,0.0000004929,0.05000,4.270
33
+ 4,self_attn.o_proj,0.0000001841,0.05000,1.432
34
+ 4,mlp.gate_proj,0.0003188560,0.05000,1.952
35
+ 4,mlp.up_proj,0.0001058938,0.05000,1.977
36
+ 4,mlp.down_proj,0.0000009877,0.05000,3.709
37
+ 5,self_attn.v_proj,0.0000004747,0.05000,4.207
38
+ 5,self_attn.k_proj,0.0000004808,0.05000,4.254
39
+ 5,self_attn.q_proj,0.0000018170,0.05000,4.288
40
+ 5,self_attn.o_proj,0.0000003358,0.05000,1.449
41
+ 5,mlp.gate_proj,0.0001046843,0.05000,2.039
42
+ 5,mlp.up_proj,0.0000414282,0.05000,2.070
43
+ 5,mlp.down_proj,0.0000013782,0.05000,3.819
44
+ 6,self_attn.v_proj,0.0000010176,0.05000,4.233
45
+ 6,self_attn.q_proj,0.0000040303,0.05000,4.253
46
+ 6,self_attn.k_proj,0.0000009895,0.05000,4.313
47
+ 6,self_attn.o_proj,0.0000006622,0.05000,1.448
48
+ 6,mlp.up_proj,0.0000503757,0.05000,2.014
49
+ 6,mlp.gate_proj,0.0001110306,0.05000,2.035
50
+ 6,mlp.down_proj,0.0000206670,0.05000,3.659
51
+ 7,self_attn.k_proj,0.0000024329,0.05000,4.064
52
+ 7,self_attn.q_proj,0.0000085931,0.05000,4.123
53
+ 7,self_attn.v_proj,0.0000019982,0.05000,4.144
54
+ 7,self_attn.o_proj,0.0000009256,0.05000,1.444
55
+ 7,mlp.gate_proj,0.0001217811,0.05000,2.071
56
+ 7,mlp.up_proj,0.0000577930,0.05000,2.092
57
+ 7,mlp.down_proj,0.0000025102,0.05000,3.710
58
+ 8,self_attn.v_proj,0.0000027738,0.05000,4.201
59
+ 8,self_attn.q_proj,0.0000114360,0.05000,4.238
60
+ 8,self_attn.k_proj,0.0000030762,0.05000,4.259
61
+ 8,self_attn.o_proj,0.0000009674,0.05000,1.412
62
+ 8,mlp.gate_proj,0.0000708923,0.05000,2.010
63
+ 8,mlp.up_proj,0.0000563776,0.05000,2.033
64
+ 8,mlp.down_proj,0.0000032682,0.05000,3.754
65
+ 9,self_attn.v_proj,0.0000035083,0.05000,4.188
66
+ 9,self_attn.k_proj,0.0000047883,0.05000,4.223
67
+ 9,self_attn.q_proj,0.0000157103,0.05000,4.262
68
+ 9,self_attn.o_proj,0.0000014919,0.05000,1.426
69
+ 9,mlp.up_proj,0.0000711674,0.05000,2.011
70
+ 9,mlp.gate_proj,0.0001187128,0.05000,2.031
71
+ 9,mlp.down_proj,0.0000035517,0.05000,3.705
72
+ 10,self_attn.k_proj,0.0000056646,0.05000,4.156
73
+ 10,self_attn.q_proj,0.0000214887,0.05000,4.234
74
+ 10,self_attn.v_proj,0.0000053920,0.05000,4.256
75
+ 10,self_attn.o_proj,0.0000015159,0.05000,1.484
76
+ 10,mlp.gate_proj,0.0000876355,0.05000,2.018
77
+ 10,mlp.up_proj,0.0000582840,0.05000,2.034
78
+ 10,mlp.down_proj,0.0000028648,0.05000,3.671
79
+ 11,self_attn.k_proj,0.0000029401,0.05000,4.225
80
+ 11,self_attn.v_proj,0.0000024203,0.05000,4.303
81
+ 11,self_attn.q_proj,0.0000103068,0.05000,4.329
82
+ 11,self_attn.o_proj,0.0000008888,0.05000,1.416
83
+ 11,mlp.gate_proj,0.0000741166,0.05000,1.917
84
+ 11,mlp.up_proj,0.0000566025,0.05000,1.941
85
+ 11,mlp.down_proj,0.0000030860,0.05000,3.775
86
+ 12,self_attn.k_proj,0.0000038472,0.05000,4.211
87
+ 12,self_attn.q_proj,0.0000137759,0.05000,4.249
88
+ 12,self_attn.v_proj,0.0000034272,0.05000,4.280
89
+ 12,self_attn.o_proj,0.0000018546,0.05000,1.426
90
+ 12,mlp.up_proj,0.0000580738,0.05000,2.071
91
+ 12,mlp.gate_proj,0.0000685725,0.05000,2.094
92
+ 12,mlp.down_proj,0.0000032841,0.05000,3.679
93
+ 13,self_attn.v_proj,0.0000025200,0.05000,4.194
94
+ 13,self_attn.q_proj,0.0000111981,0.05000,4.230
95
+ 13,self_attn.k_proj,0.0000030891,0.05000,4.257
96
+ 13,self_attn.o_proj,0.0000013833,0.05000,1.412
97
+ 13,mlp.gate_proj,0.0000692568,0.05000,1.954
98
+ 13,mlp.up_proj,0.0000647811,0.05000,1.973
99
+ 13,mlp.down_proj,0.0000043834,0.05000,3.687
100
+ 14,self_attn.q_proj,0.0000175752,0.05000,4.123
101
+ 14,self_attn.k_proj,0.0000045994,0.05000,4.169
102
+ 14,self_attn.v_proj,0.0000043383,0.05000,4.177
103
+ 14,self_attn.o_proj,0.0000019435,0.05000,1.440
104
+ 14,mlp.up_proj,0.0000720079,0.05000,1.933
105
+ 14,mlp.gate_proj,0.0000766432,0.05000,1.958
106
+ 14,mlp.down_proj,0.0000047673,0.05000,3.682
107
+ 15,self_attn.k_proj,0.0000054351,0.05000,4.116
108
+ 15,self_attn.v_proj,0.0000046581,0.05000,4.200
109
+ 15,self_attn.q_proj,0.0000211606,0.05000,4.223
110
+ 15,self_attn.o_proj,0.0000018997,0.05000,1.429
111
+ 15,mlp.gate_proj,0.0000754266,0.05000,1.990
112
+ 15,mlp.up_proj,0.0000740122,0.05000,2.015
113
+ 15,mlp.down_proj,0.0000053597,0.05000,3.737
114
+ 16,self_attn.q_proj,0.0000386672,0.05000,4.067
115
+ 16,self_attn.k_proj,0.0000093139,0.05000,4.130
116
+ 16,self_attn.v_proj,0.0000103499,0.05000,4.155
117
+ 16,self_attn.o_proj,0.0000026699,0.05000,1.403
118
+ 16,mlp.up_proj,0.0000855751,0.05000,2.041
119
+ 16,mlp.gate_proj,0.0000930978,0.05000,2.061
120
+ 16,mlp.down_proj,0.0000275581,0.05000,3.699
121
+ 17,self_attn.q_proj,0.0000392438,0.05000,4.191
122
+ 17,self_attn.k_proj,0.0000084165,0.05000,4.234
123
+ 17,self_attn.v_proj,0.0000087300,0.05000,4.253
124
+ 17,self_attn.o_proj,0.0000026301,0.05000,1.434
125
+ 17,mlp.gate_proj,0.0000779856,0.05000,2.005
126
+ 17,mlp.up_proj,0.0000724824,0.05000,2.024
127
+ 17,mlp.down_proj,0.0000052021,0.05000,3.706
128
+ 18,self_attn.q_proj,0.0000359746,0.05000,4.202
129
+ 18,self_attn.v_proj,0.0000089155,0.05000,4.239
130
+ 18,self_attn.k_proj,0.0000086172,0.05000,4.263
131
+ 18,self_attn.o_proj,0.0000026857,0.05000,1.403
132
+ 18,mlp.gate_proj,0.0000826436,0.05000,1.962
133
+ 18,mlp.up_proj,0.0000786649,0.05000,1.977
134
+ 18,mlp.down_proj,0.0000061125,0.05000,3.797
135
+ 19,self_attn.k_proj,0.0000154321,0.05000,4.153
136
+ 19,self_attn.v_proj,0.0000145288,0.05000,4.226
137
+ 19,self_attn.q_proj,0.0000669893,0.05000,4.257
138
+ 19,self_attn.o_proj,0.0000072475,0.05000,1.420
139
+ 19,mlp.up_proj,0.0000763673,0.05000,1.958
140
+ 19,mlp.gate_proj,0.0000795653,0.05000,1.983
141
+ 19,mlp.down_proj,0.0000063999,0.05000,3.722
142
+ 20,self_attn.k_proj,0.0000139938,0.05000,4.166
143
+ 20,self_attn.v_proj,0.0000126008,0.05000,4.255
144
+ 20,self_attn.q_proj,0.0000648620,0.05000,4.270
145
+ 20,self_attn.o_proj,0.0000058633,0.05000,1.471
146
+ 20,mlp.gate_proj,0.0000927177,0.05000,1.983
147
+ 20,mlp.up_proj,0.0000909301,0.05000,2.005
148
+ 20,mlp.down_proj,0.0000085226,0.05000,3.687
149
+ 21,self_attn.v_proj,0.0000157023,0.05000,4.200
150
+ 21,self_attn.q_proj,0.0000732686,0.05000,4.246
151
+ 21,self_attn.k_proj,0.0000162576,0.05000,4.279
152
+ 21,self_attn.o_proj,0.0000064938,0.05000,1.451
153
+ 21,mlp.up_proj,0.0000938207,0.05000,2.061
154
+ 21,mlp.gate_proj,0.0000953118,0.05000,2.069
155
+ 21,mlp.down_proj,0.0000094180,0.05000,3.670
156
+ 22,self_attn.v_proj,0.0000290647,0.05000,4.262
157
+ 22,self_attn.q_proj,0.0001376551,0.05000,4.291
158
+ 22,self_attn.k_proj,0.0000316722,0.05000,4.321
159
+ 22,self_attn.o_proj,0.0000139498,0.05000,1.408
160
+ 22,mlp.up_proj,0.0001028317,0.05000,1.960
161
+ 22,mlp.gate_proj,0.0001130452,0.05000,1.982
162
+ 22,mlp.down_proj,0.0000150741,0.05000,3.654
163
+ 23,self_attn.q_proj,0.0001271593,0.05000,4.130
164
+ 23,self_attn.k_proj,0.0000288629,0.05000,4.167
165
+ 23,self_attn.v_proj,0.0000286624,0.05000,4.196
166
+ 23,self_attn.o_proj,0.0000121074,0.05000,1.435
167
+ 23,mlp.up_proj,0.0001123853,0.05000,1.978
168
+ 23,mlp.gate_proj,0.0001318009,0.05000,2.002
169
+ 23,mlp.down_proj,0.0000158443,0.05000,3.680
170
+ 24,self_attn.k_proj,0.0000445231,0.05000,4.180
171
+ 24,self_attn.v_proj,0.0000517156,0.05000,4.218
172
+ 24,self_attn.q_proj,0.0001876024,0.05000,4.237
173
+ 24,self_attn.o_proj,0.0000197756,0.05000,1.443
174
+ 24,mlp.gate_proj,0.0001390896,0.05000,2.028
175
+ 24,mlp.up_proj,0.0001163446,0.05000,2.052
176
+ 24,mlp.down_proj,0.0000183784,0.05000,3.721
177
+ 25,self_attn.k_proj,0.0000309183,0.05000,4.098
178
+ 25,self_attn.v_proj,0.0000312976,0.05000,4.159
179
+ 25,self_attn.q_proj,0.0001220958,0.05000,4.188
180
+ 25,self_attn.o_proj,0.0000078953,0.05000,1.390
181
+ 25,mlp.gate_proj,0.0001551162,0.05000,1.971
182
+ 25,mlp.up_proj,0.0001274080,0.05000,1.993
183
+ 25,mlp.down_proj,0.0000235486,0.05000,3.673
184
+ 26,self_attn.v_proj,0.0000490624,0.05000,4.113
185
+ 26,self_attn.k_proj,0.0000464167,0.05000,4.161
186
+ 26,self_attn.q_proj,0.0001926069,0.05000,4.185
187
+ 26,self_attn.o_proj,0.0000066098,0.05000,1.387
188
+ 26,mlp.gate_proj,0.0001770693,0.05000,1.976
189
+ 26,mlp.up_proj,0.0001508838,0.05000,1.975
190
+ 26,mlp.down_proj,0.0000279487,0.05000,3.647
191
+ 27,self_attn.k_proj,0.0000556290,0.05000,4.077
192
+ 27,self_attn.q_proj,0.0002357567,0.05000,4.141
193
+ 27,self_attn.v_proj,0.0000654981,0.05000,4.174
194
+ 27,self_attn.o_proj,0.0000107555,0.05000,1.416
195
+ 27,mlp.gate_proj,0.0001936240,0.05000,1.989
196
+ 27,mlp.up_proj,0.0001708205,0.05000,2.008
197
+ 27,mlp.down_proj,0.0000438858,0.05000,3.659
198
+ 28,self_attn.k_proj,0.0000705007,0.05000,4.196
199
+ 28,self_attn.v_proj,0.0000715719,0.05000,4.249
200
+ 28,self_attn.q_proj,0.0002718625,0.05000,4.266
201
+ 28,self_attn.o_proj,0.0000191249,0.05000,1.394
202
+ 28,mlp.gate_proj,0.0002182023,0.05000,2.015
203
+ 28,mlp.up_proj,0.0002019346,0.05000,2.031
204
+ 28,mlp.down_proj,0.0000732407,0.05000,3.797
205
+ 29,self_attn.k_proj,0.0001816301,0.05000,4.092
206
+ 29,self_attn.q_proj,0.0007383766,0.05000,4.169
207
+ 29,self_attn.v_proj,0.0002443326,0.05000,4.195
208
+ 29,self_attn.o_proj,0.0000448207,0.05000,1.481
209
+ 29,mlp.up_proj,0.0002562076,0.05000,1.942
210
+ 29,mlp.gate_proj,0.0002613911,0.05000,1.958
211
+ 29,mlp.down_proj,0.0000526463,0.05000,3.700
212
+ 30,self_attn.v_proj,0.0002320336,0.05000,4.041
213
+ 30,self_attn.k_proj,0.0001986943,0.05000,4.134
214
+ 30,self_attn.q_proj,0.0007668271,0.05000,4.142
215
+ 30,self_attn.o_proj,0.0000478573,0.05000,1.459
216
+ 30,mlp.gate_proj,0.0002683302,0.05000,1.898
217
+ 30,mlp.up_proj,0.0002727380,0.05000,1.920
218
+ 30,mlp.down_proj,0.0000663494,0.05000,3.662
219
+ 31,self_attn.q_proj,0.0010675234,0.05000,4.039
220
+ 31,self_attn.k_proj,0.0003015756,0.05000,4.116
221
+ 31,self_attn.v_proj,0.0003918752,0.05000,4.162
222
+ 31,self_attn.o_proj,0.0000612622,0.05000,1.394
223
+ 31,mlp.up_proj,0.0002870251,0.05000,2.027
224
+ 31,mlp.gate_proj,0.0002682364,0.05000,2.032
225
+ 31,mlp.down_proj,0.0000840589,0.05000,3.717
226
+ 32,self_attn.q_proj,0.0016584295,0.05000,4.132
227
+ 32,self_attn.v_proj,0.0006243234,0.05000,4.233
228
+ 32,self_attn.k_proj,0.0004182623,0.05000,4.253
229
+ 32,self_attn.o_proj,0.0001057172,0.05000,1.428
230
+ 32,mlp.up_proj,0.0003151746,0.05000,1.983
231
+ 32,mlp.gate_proj,0.0002878729,0.05000,2.006
232
+ 32,mlp.down_proj,0.0001186975,0.05000,3.678
233
+ 33,self_attn.v_proj,0.0016297800,0.05000,4.071
234
+ 33,self_attn.k_proj,0.0008791048,0.05000,4.100
235
+ 33,self_attn.q_proj,0.0041213692,0.05000,4.145
236
+ 33,self_attn.o_proj,0.0001997127,0.05000,1.434
237
+ 33,mlp.gate_proj,0.0003185476,0.05000,1.865
238
+ 33,mlp.up_proj,0.0003529478,0.05000,1.881
239
+ 33,mlp.down_proj,0.0001623388,0.05000,3.669
240
+ 34,self_attn.q_proj,0.0027944569,0.05000,4.110
241
+ 34,self_attn.k_proj,0.0006274358,0.05000,4.194
242
+ 34,self_attn.v_proj,0.0009633384,0.05000,4.209
243
+ 34,self_attn.o_proj,0.0002593519,0.05000,1.436
244
+ 34,mlp.up_proj,0.0004195679,0.05000,1.947
245
+ 34,mlp.gate_proj,0.0004068270,0.05000,1.970
246
+ 34,mlp.down_proj,0.0003025797,0.05000,3.683
247
+ 35,self_attn.v_proj,0.0003734207,0.05000,4.091
248
+ 35,self_attn.k_proj,0.0003153036,0.05000,4.159
249
+ 35,self_attn.q_proj,0.0011829730,0.05000,4.173
250
+ 35,self_attn.o_proj,0.0001886140,0.05000,1.407
251
+ 35,mlp.gate_proj,0.0009205786,0.05000,1.941
252
+ 35,mlp.up_proj,0.0009714007,0.05000,1.939
253
+ 35,mlp.down_proj,0.0008959190,0.05000,3.653
Models/70/quantize_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bits": 4,
3
+ "group_size": 128,
4
+ "desc_act": false,
5
+ "lm_head": false,
6
+ "method": "gptq",
7
+ "quant_method": "gptq",
8
+ "format": "gptq",
9
+ "checkpoint_format": "gptq",
10
+ "pack_dtype": "int32",
11
+ "meta": {
12
+ "quantizer": [
13
+ "gptqmodel:6.0.0"
14
+ ],
15
+ "uri": "https://github.com/modelcloud/gptqmodel",
16
+ "damp_percent": 0.05,
17
+ "damp_auto_increment": 0.01,
18
+ "static_groups": false,
19
+ "true_sequential": true,
20
+ "mse": 0.0,
21
+ "gptaq": null,
22
+ "foem": null,
23
+ "act_group_aware": true,
24
+ "fallback": {
25
+ "strategy": "rtn",
26
+ "threshold": "0.5%",
27
+ "smooth": null
28
+ },
29
+ "offload_to_disk": true,
30
+ "offload_to_disk_path": "./gptqmodel_offload/hkydshso-ldwypfzp/",
31
+ "pack_impl": "cpu",
32
+ "gc_mode": "interval",
33
+ "wait_for_submodule_finalizers": false,
34
+ "auto_forward_data_parallel": true,
35
+ "vram_strategy": "exclusive",
36
+ "mock_quantization": false,
37
+ "hessian": {
38
+ "chunk_size": null,
39
+ "chunk_bytes": null,
40
+ "staging_dtype": "float32"
41
+ }
42
+ },
43
+ "sym": true
44
+ }
Models/70/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:749876d361036526939013c5243386793a5ccf69b68004faff0bab4b9c156991
3
+ size 11422925
Models/70/tokenizer_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": null,
5
+ "clean_up_tokenization_spaces": false,
6
+ "eos_token": "<|im_end|>",
7
+ "errors": "replace",
8
+ "extra_special_tokens": [
9
+ "<|im_start|>",
10
+ "<|im_end|>",
11
+ "<|object_ref_start|>",
12
+ "<|object_ref_end|>",
13
+ "<|box_start|>",
14
+ "<|box_end|>",
15
+ "<|quad_start|>",
16
+ "<|quad_end|>",
17
+ "<|vision_start|>",
18
+ "<|vision_end|>",
19
+ "<|vision_pad|>",
20
+ "<|image_pad|>",
21
+ "<|video_pad|>"
22
+ ],
23
+ "is_local": false,
24
+ "model_max_length": 131072,
25
+ "pad_token": "<|im_end|>",
26
+ "split_special_tokens": false,
27
+ "tokenizer_class": "Qwen2Tokenizer",
28
+ "unk_token": null
29
+ }