Hugging Face's logo

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

Safetensors
Model card Files
xet
 Community
Abdohaaland commited on
Commit
0769b7b
·
verified ·
1 Parent(s): 9ddf1f7

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 +113 -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 +42 -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 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 151643,
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
+ "failsafe": {
73
+ "smooth": null,
74
+ "strategy": "rtn",
75
+ "threshold": "0.5%"
76
+ },
77
+ "gc_mode": "interval",
78
+ "gptaq": null,
79
+ "hessian": {
80
+ "chunk_bytes": null,
81
+ "chunk_size": null,
82
+ "staging_dtype": "float32"
83
+ },
84
+ "mock_quantization": false,
85
+ "mse": 0.0,
86
+ "offload_to_disk": true,
87
+ "offload_to_disk_path": "./gptqmodel_offload/llkguovp-bhlrefln/",
88
+ "pack_impl": "cpu",
89
+ "quantizer": [
90
+ "gptqmodel:5.8.0"
91
+ ],
92
+ "static_groups": false,
93
+ "true_sequential": true,
94
+ "uri": "https://github.com/modelcloud/gptqmodel",
95
+ "vram_strategy": "exclusive",
96
+ "wait_for_submodule_finalizers": false
97
+ },
98
+ "pack_dtype": "int32",
99
+ "quant_method": "gptq",
100
+ "sym": true
101
+ },
102
+ "rms_norm_eps": 1e-06,
103
+ "rope_parameters": {
104
+ "rope_theta": 1000000,
105
+ "rope_type": "default"
106
+ },
107
+ "sliding_window": null,
108
+ "tie_word_embeddings": true,
109
+ "transformers_version": "5.2.0",
110
+ "use_cache": true,
111
+ "use_sliding_window": false,
112
+ "vocab_size": 151936
113
+ }
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": "5.2.0"
12
+ }
Models/70/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9c6056ed01fd85836da8cb9e90bc8999a81496c7a8849d6fb39d1630656d5a5
3
+ size 2669888992
Models/70/quant_log.csv ADDED
@@ -0,0 +1,253 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ layer,module,loss,samples,damp,time
2
+ 0,self_attn.q_proj,0.0000003104,0.05000,4.905
3
+ 0,self_attn.v_proj,0.0000000719,0.05000,4.994
4
+ 0,self_attn.k_proj,0.0000000848,0.05000,5.018
5
+ 0,self_attn.o_proj,0.0000001642,0.05000,1.457
6
+ 0,mlp.up_proj,0.0000166251,0.05000,1.845
7
+ 0,mlp.gate_proj,0.0000194163,0.05000,1.875
8
+ 0,mlp.down_proj,0.0000009856,0.05000,3.691
9
+ 1,self_attn.k_proj,0.0000001349,0.05000,4.619
10
+ 1,self_attn.q_proj,0.0000004949,0.05000,4.702
11
+ 1,self_attn.v_proj,0.0000001330,0.05000,4.728
12
+ 1,self_attn.o_proj,0.0000001566,0.05000,1.611
13
+ 1,mlp.gate_proj,0.0005942452,0.05000,2.445
14
+ 1,mlp.up_proj,0.0002664099,0.05000,2.507
15
+ 1,mlp.down_proj,0.0000016856,0.05000,3.879
16
+ 2,self_attn.k_proj,0.0000003772,0.05000,4.616
17
+ 2,self_attn.q_proj,0.0000012779,0.05000,4.651
18
+ 2,self_attn.v_proj,0.0000003293,0.05000,4.678
19
+ 2,self_attn.o_proj,0.0000002073,0.05000,1.550
20
+ 2,mlp.gate_proj,0.0010680482,0.05000,2.444
21
+ 2,mlp.up_proj,0.0008312045,0.05000,2.475
22
+ 2,mlp.down_proj,0.0000015937,0.05000,3.758
23
+ 3,self_attn.k_proj,0.0000007141,0.05000,4.042
24
+ 3,self_attn.q_proj,0.0000027178,0.05000,4.098
25
+ 3,self_attn.v_proj,0.0000006610,0.05000,4.133
26
+ 3,self_attn.o_proj,0.0000004343,0.05000,1.443
27
+ 3,mlp.up_proj,0.0004659894,0.05000,1.978
28
+ 3,mlp.gate_proj,0.0009418213,0.05000,1.998
29
+ 3,mlp.down_proj,0.0000026344,0.05000,3.723
30
+ 4,self_attn.v_proj,0.0000011467,0.05000,4.135
31
+ 4,self_attn.k_proj,0.0000012765,0.05000,4.217
32
+ 4,self_attn.q_proj,0.0000045529,0.05000,4.233
33
+ 4,self_attn.o_proj,0.0000005542,0.05000,1.409
34
+ 4,mlp.gate_proj,0.0008452089,0.05000,2.026
35
+ 4,mlp.up_proj,0.0002988149,0.05000,2.043
36
+ 4,mlp.down_proj,0.0000032080,0.05000,3.691
37
+ 5,self_attn.k_proj,0.0000012628,0.05000,4.213
38
+ 5,self_attn.v_proj,0.0000012745,0.05000,4.269
39
+ 5,self_attn.q_proj,0.0000048576,0.05000,4.303
40
+ 5,self_attn.o_proj,0.0000010330,0.05000,1.439
41
+ 5,mlp.up_proj,0.0001115446,0.05000,2.013
42
+ 5,mlp.gate_proj,0.0002728422,0.05000,2.022
43
+ 5,mlp.down_proj,0.0000040210,0.05000,3.732
44
+ 6,self_attn.k_proj,0.0000024965,0.05000,4.120
45
+ 6,self_attn.q_proj,0.0000103244,0.05000,4.136
46
+ 6,self_attn.v_proj,0.0000026516,0.05000,4.181
47
+ 6,self_attn.o_proj,0.0000022328,0.05000,1.425
48
+ 6,mlp.up_proj,0.0001320891,0.05000,1.990
49
+ 6,mlp.gate_proj,0.0002955694,0.05000,1.999
50
+ 6,mlp.down_proj,0.0000908862,0.05000,3.725
51
+ 7,self_attn.q_proj,0.0000219465,0.05000,4.214
52
+ 7,self_attn.k_proj,0.0000061061,0.05000,4.320
53
+ 7,self_attn.v_proj,0.0000053100,0.05000,4.340
54
+ 7,self_attn.o_proj,0.0000029166,0.05000,1.409
55
+ 7,mlp.gate_proj,0.0003294983,0.05000,1.882
56
+ 7,mlp.up_proj,0.0001593253,0.05000,1.897
57
+ 7,mlp.down_proj,0.0000070525,0.05000,3.790
58
+ 8,self_attn.v_proj,0.0000074580,0.05000,4.248
59
+ 8,self_attn.q_proj,0.0000303084,0.05000,4.315
60
+ 8,self_attn.k_proj,0.0000081583,0.05000,4.334
61
+ 8,self_attn.o_proj,0.0000036968,0.05000,1.403
62
+ 8,mlp.up_proj,0.0001478816,0.05000,1.953
63
+ 8,mlp.gate_proj,0.0001890822,0.05000,1.967
64
+ 8,mlp.down_proj,0.0000118579,0.05000,3.753
65
+ 9,self_attn.v_proj,0.0000095992,0.05000,4.190
66
+ 9,self_attn.q_proj,0.0000413045,0.05000,4.249
67
+ 9,self_attn.k_proj,0.0000121445,0.05000,4.268
68
+ 9,self_attn.o_proj,0.0000053125,0.05000,1.515
69
+ 9,mlp.gate_proj,0.0003300729,0.05000,1.967
70
+ 9,mlp.up_proj,0.0001974271,0.05000,1.986
71
+ 9,mlp.down_proj,0.0000125098,0.05000,3.755
72
+ 10,self_attn.k_proj,0.0000153336,0.05000,4.206
73
+ 10,self_attn.q_proj,0.0000581782,0.05000,4.224
74
+ 10,self_attn.v_proj,0.0000151345,0.05000,4.264
75
+ 10,self_attn.o_proj,0.0000061027,0.05000,1.396
76
+ 10,mlp.up_proj,0.0001707550,0.05000,1.915
77
+ 10,mlp.gate_proj,0.0002543201,0.05000,1.933
78
+ 10,mlp.down_proj,0.0000103128,0.05000,3.691
79
+ 11,self_attn.k_proj,0.0000080204,0.05000,4.166
80
+ 11,self_attn.q_proj,0.0000286084,0.05000,4.207
81
+ 11,self_attn.v_proj,0.0000070372,0.05000,4.228
82
+ 11,self_attn.o_proj,0.0000043562,0.05000,1.416
83
+ 11,mlp.up_proj,0.0001597086,0.05000,1.959
84
+ 11,mlp.gate_proj,0.0002079164,0.05000,1.979
85
+ 11,mlp.down_proj,0.0000100665,0.05000,3.695
86
+ 12,self_attn.q_proj,0.0000361063,0.05000,4.147
87
+ 12,self_attn.k_proj,0.0000098947,0.05000,4.237
88
+ 12,self_attn.v_proj,0.0000093602,0.05000,4.269
89
+ 12,self_attn.o_proj,0.0000063775,0.05000,1.406
90
+ 12,mlp.up_proj,0.0001564673,0.05000,1.963
91
+ 12,mlp.gate_proj,0.0001846403,0.05000,1.958
92
+ 12,mlp.down_proj,0.0000100620,0.05000,3.720
93
+ 13,self_attn.q_proj,0.0000280880,0.05000,4.138
94
+ 13,self_attn.k_proj,0.0000076636,0.05000,4.192
95
+ 13,self_attn.v_proj,0.0000065074,0.05000,4.205
96
+ 13,self_attn.o_proj,0.0000054739,0.05000,1.425
97
+ 13,mlp.up_proj,0.0001643997,0.05000,2.025
98
+ 13,mlp.gate_proj,0.0001773453,0.05000,2.042
99
+ 13,mlp.down_proj,0.0000121140,0.05000,3.696
100
+ 14,self_attn.q_proj,0.0000446386,0.05000,4.120
101
+ 14,self_attn.v_proj,0.0000111164,0.05000,4.194
102
+ 14,self_attn.k_proj,0.0000117079,0.05000,4.227
103
+ 14,self_attn.o_proj,0.0000080355,0.05000,1.390
104
+ 14,mlp.up_proj,0.0001662609,0.05000,1.928
105
+ 14,mlp.gate_proj,0.0001788644,0.05000,1.951
106
+ 14,mlp.down_proj,0.0000115574,0.05000,3.707
107
+ 15,self_attn.k_proj,0.0000127820,0.05000,3.983
108
+ 15,self_attn.v_proj,0.0000108480,0.05000,4.041
109
+ 15,self_attn.q_proj,0.0000485437,0.05000,4.066
110
+ 15,self_attn.o_proj,0.0000079289,0.05000,1.396
111
+ 15,mlp.up_proj,0.0001686920,0.05000,1.938
112
+ 15,mlp.gate_proj,0.0001739771,0.05000,1.951
113
+ 15,mlp.down_proj,0.0000120265,0.05000,3.730
114
+ 16,self_attn.v_proj,0.0000226708,0.05000,4.071
115
+ 16,self_attn.q_proj,0.0000885504,0.05000,4.176
116
+ 16,self_attn.k_proj,0.0000220067,0.05000,4.193
117
+ 16,self_attn.o_proj,0.0000097078,0.05000,1.403
118
+ 16,mlp.up_proj,0.0001986332,0.05000,1.905
119
+ 16,mlp.gate_proj,0.0002191926,0.05000,1.925
120
+ 16,mlp.down_proj,0.0001195219,0.05000,3.697
121
+ 17,self_attn.v_proj,0.0000192036,0.05000,4.160
122
+ 17,self_attn.q_proj,0.0000859028,0.05000,4.196
123
+ 17,self_attn.k_proj,0.0000187684,0.05000,4.219
124
+ 17,self_attn.o_proj,0.0000104221,0.05000,1.429
125
+ 17,mlp.up_proj,0.0001920366,0.05000,1.979
126
+ 17,mlp.gate_proj,0.0002083335,0.05000,1.997
127
+ 17,mlp.down_proj,0.0000144278,0.05000,3.702
128
+ 18,self_attn.q_proj,0.0000918104,0.05000,3.970
129
+ 18,self_attn.k_proj,0.0000223279,0.05000,4.084
130
+ 18,self_attn.v_proj,0.0000228247,0.05000,4.105
131
+ 18,self_attn.o_proj,0.0000120030,0.05000,1.408
132
+ 18,mlp.up_proj,0.0002062233,0.05000,1.959
133
+ 18,mlp.gate_proj,0.0002194696,0.05000,1.968
134
+ 18,mlp.down_proj,0.0000170803,0.05000,3.642
135
+ 19,self_attn.q_proj,0.0001786820,0.05000,4.068
136
+ 19,self_attn.v_proj,0.0000392877,0.05000,4.147
137
+ 19,self_attn.k_proj,0.0000417310,0.05000,4.178
138
+ 19,self_attn.o_proj,0.0000172448,0.05000,1.420
139
+ 19,mlp.up_proj,0.0002172751,0.05000,2.036
140
+ 19,mlp.gate_proj,0.0002297170,0.05000,2.057
141
+ 19,mlp.down_proj,0.0000206614,0.05000,3.705
142
+ 20,self_attn.q_proj,0.0001758999,0.05000,4.152
143
+ 20,self_attn.v_proj,0.0000377428,0.05000,4.207
144
+ 20,self_attn.k_proj,0.0000383674,0.05000,4.233
145
+ 20,self_attn.o_proj,0.0000194928,0.05000,1.407
146
+ 20,mlp.gate_proj,0.0002446541,0.05000,1.948
147
+ 20,mlp.up_proj,0.0002372765,0.05000,1.968
148
+ 20,mlp.down_proj,0.0000223872,0.05000,3.734
149
+ 21,self_attn.v_proj,0.0000462733,0.05000,4.032
150
+ 21,self_attn.k_proj,0.0000457615,0.05000,4.107
151
+ 21,self_attn.q_proj,0.0002003734,0.05000,4.130
152
+ 21,self_attn.o_proj,0.0000193219,0.05000,1.382
153
+ 21,mlp.up_proj,0.0002598240,0.05000,1.994
154
+ 21,mlp.gate_proj,0.0002718615,0.05000,2.013
155
+ 21,mlp.down_proj,0.0000280902,0.05000,3.820
156
+ 22,self_attn.v_proj,0.0000893241,0.05000,4.115
157
+ 22,self_attn.k_proj,0.0000910953,0.05000,4.191
158
+ 22,self_attn.q_proj,0.0003968653,0.05000,4.210
159
+ 22,self_attn.o_proj,0.0000352733,0.05000,1.394
160
+ 22,mlp.up_proj,0.0002868396,0.05000,1.933
161
+ 22,mlp.gate_proj,0.0003197921,0.05000,1.949
162
+ 22,mlp.down_proj,0.0000435962,0.05000,3.718
163
+ 23,self_attn.k_proj,0.0000845952,0.05000,4.110
164
+ 23,self_attn.q_proj,0.0003595412,0.05000,4.206
165
+ 23,self_attn.v_proj,0.0000884456,0.05000,4.234
166
+ 23,self_attn.o_proj,0.0000378814,0.05000,1.389
167
+ 23,mlp.gate_proj,0.0003853287,0.05000,1.959
168
+ 23,mlp.up_proj,0.0003278120,0.05000,1.955
169
+ 23,mlp.down_proj,0.0000534777,0.05000,3.675
170
+ 24,self_attn.k_proj,0.0001351002,0.05000,4.143
171
+ 24,self_attn.v_proj,0.0001678118,0.05000,4.186
172
+ 24,self_attn.q_proj,0.0005597780,0.05000,4.222
173
+ 24,self_attn.o_proj,0.0000466474,0.05000,1.423
174
+ 24,mlp.up_proj,0.0003403691,0.05000,1.974
175
+ 24,mlp.gate_proj,0.0004069859,0.05000,1.992
176
+ 24,mlp.down_proj,0.0000601208,0.05000,3.689
177
+ 25,self_attn.q_proj,0.0003528415,0.05000,4.221
178
+ 25,self_attn.k_proj,0.0000905896,0.05000,4.284
179
+ 25,self_attn.v_proj,0.0000948952,0.05000,4.304
180
+ 25,self_attn.o_proj,0.0000231992,0.05000,1.395
181
+ 25,mlp.up_proj,0.0003582715,0.05000,2.032
182
+ 25,mlp.gate_proj,0.0004356605,0.05000,2.049
183
+ 25,mlp.down_proj,0.0000689074,0.05000,3.674
184
+ 26,self_attn.k_proj,0.0001325942,0.05000,4.171
185
+ 26,self_attn.v_proj,0.0001480613,0.05000,4.257
186
+ 26,self_attn.q_proj,0.0005470764,0.05000,4.309
187
+ 26,self_attn.o_proj,0.0000218077,0.05000,1.398
188
+ 26,mlp.gate_proj,0.0004783157,0.05000,1.944
189
+ 26,mlp.up_proj,0.0004074046,0.05000,1.974
190
+ 26,mlp.down_proj,0.0000757961,0.05000,3.774
191
+ 27,self_attn.v_proj,0.0001919626,0.05000,4.095
192
+ 27,self_attn.q_proj,0.0006371495,0.05000,4.127
193
+ 27,self_attn.k_proj,0.0001516362,0.05000,4.139
194
+ 27,self_attn.o_proj,0.0000294259,0.05000,1.432
195
+ 27,mlp.gate_proj,0.0005036255,0.05000,1.895
196
+ 27,mlp.up_proj,0.0004426054,0.05000,1.907
197
+ 27,mlp.down_proj,0.0001096823,0.05000,3.705
198
+ 28,self_attn.v_proj,0.0002048029,0.05000,4.179
199
+ 28,self_attn.q_proj,0.0007217175,0.05000,4.261
200
+ 28,self_attn.k_proj,0.0001894458,0.05000,4.277
201
+ 28,self_attn.o_proj,0.0000451533,0.05000,1.456
202
+ 28,mlp.up_proj,0.0004963349,0.05000,1.984
203
+ 28,mlp.gate_proj,0.0005422109,0.05000,1.999
204
+ 28,mlp.down_proj,0.0001637474,0.05000,3.662
205
+ 29,self_attn.q_proj,0.0017744177,0.05000,4.176
206
+ 29,self_attn.v_proj,0.0006037154,0.05000,4.219
207
+ 29,self_attn.k_proj,0.0004375837,0.05000,4.246
208
+ 29,self_attn.o_proj,0.0000526574,0.05000,1.403
209
+ 29,mlp.gate_proj,0.0005959541,0.05000,2.040
210
+ 29,mlp.up_proj,0.0005792906,0.05000,2.057
211
+ 29,mlp.down_proj,0.0001176740,0.05000,3.716
212
+ 30,self_attn.q_proj,0.0017855356,0.05000,4.072
213
+ 30,self_attn.k_proj,0.0004658168,0.05000,4.109
214
+ 30,self_attn.v_proj,0.0005561837,0.05000,4.118
215
+ 30,self_attn.o_proj,0.0000770569,0.05000,1.385
216
+ 30,mlp.up_proj,0.0005726860,0.05000,1.973
217
+ 30,mlp.gate_proj,0.0005691606,0.05000,1.992
218
+ 30,mlp.down_proj,0.0001377208,0.05000,3.679
219
+ 31,self_attn.k_proj,0.0006738857,0.05000,4.074
220
+ 31,self_attn.q_proj,0.0023534875,0.05000,4.180
221
+ 31,self_attn.v_proj,0.0009032022,0.05000,4.202
222
+ 31,self_attn.o_proj,0.0000924383,0.05000,1.398
223
+ 31,mlp.gate_proj,0.0005124867,0.05000,2.077
224
+ 31,mlp.up_proj,0.0005424235,0.05000,2.092
225
+ 31,mlp.down_proj,0.0001507833,0.05000,3.702
226
+ 32,self_attn.q_proj,0.0033660829,0.05000,4.254
227
+ 32,self_attn.v_proj,0.0012981689,0.05000,4.251
228
+ 32,self_attn.k_proj,0.0008497750,0.05000,4.311
229
+ 32,self_attn.o_proj,0.0001307035,0.05000,1.413
230
+ 32,mlp.up_proj,0.0005475206,0.05000,1.964
231
+ 32,mlp.gate_proj,0.0005065469,0.05000,1.982
232
+ 32,mlp.down_proj,0.0001925587,0.05000,3.766
233
+ 33,self_attn.q_proj,0.0081515312,0.05000,4.143
234
+ 33,self_attn.k_proj,0.0017276332,0.05000,4.244
235
+ 33,self_attn.v_proj,0.0032664668,0.05000,4.253
236
+ 33,self_attn.o_proj,0.0002568730,0.05000,1.422
237
+ 33,mlp.up_proj,0.0006197004,0.05000,1.991
238
+ 33,mlp.gate_proj,0.0005619661,0.05000,2.013
239
+ 33,mlp.down_proj,0.0002740641,0.05000,3.756
240
+ 34,self_attn.k_proj,0.0012690087,0.05000,4.134
241
+ 34,self_attn.q_proj,0.0056529627,0.05000,4.188
242
+ 34,self_attn.v_proj,0.0020060930,0.05000,4.221
243
+ 34,self_attn.o_proj,0.0002881256,0.05000,1.424
244
+ 34,mlp.up_proj,0.0007620189,0.05000,1.904
245
+ 34,mlp.gate_proj,0.0007313601,0.05000,1.906
246
+ 34,mlp.down_proj,0.0005008702,0.05000,3.746
247
+ 35,self_attn.k_proj,0.0005734052,0.05000,4.103
248
+ 35,self_attn.v_proj,0.0006820257,0.05000,4.200
249
+ 35,self_attn.q_proj,0.0021040146,0.05000,4.217
250
+ 35,self_attn.o_proj,0.0002867113,0.05000,1.422
251
+ 35,mlp.up_proj,0.0022627204,0.05000,1.957
252
+ 35,mlp.gate_proj,0.0019303147,0.05000,1.969
253
+ 35,mlp.down_proj,0.0015434768,0.05000,3.715
Models/70/quantize_config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bits": 4,
3
+ "group_size": 128,
4
+ "desc_act": false,
5
+ "lm_head": false,
6
+ "quant_method": "gptq",
7
+ "checkpoint_format": "gptq",
8
+ "pack_dtype": "int32",
9
+ "meta": {
10
+ "quantizer": [
11
+ "gptqmodel:5.8.0"
12
+ ],
13
+ "uri": "https://github.com/modelcloud/gptqmodel",
14
+ "damp_percent": 0.05,
15
+ "damp_auto_increment": 0.01,
16
+ "static_groups": false,
17
+ "true_sequential": true,
18
+ "mse": 0.0,
19
+ "gptaq": null,
20
+ "act_group_aware": true,
21
+ "failsafe": {
22
+ "strategy": "rtn",
23
+ "threshold": "0.5%",
24
+ "smooth": null
25
+ },
26
+ "offload_to_disk": true,
27
+ "offload_to_disk_path": "./gptqmodel_offload/llkguovp-bhlrefln/",
28
+ "pack_impl": "cpu",
29
+ "mock_quantization": false,
30
+ "gc_mode": "interval",
31
+ "wait_for_submodule_finalizers": false,
32
+ "auto_forward_data_parallel": true,
33
+ "hessian": {
34
+ "chunk_size": null,
35
+ "chunk_bytes": null,
36
+ "staging_dtype": "float32"
37
+ },
38
+ "vram_strategy": "exclusive"
39
+ },
40
+ "sym": true,
41
+ "format": "gptq"
42
+ }
Models/70/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7029094cd70eca33e2f5d6837051bd1b63789ebde3c05bcce93b0fb31c094a85
3
+ size 11422928
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": "<|endoftext|>",
26
+ "split_special_tokens": false,
27
+ "tokenizer_class": "Qwen2Tokenizer",
28
+ "unk_token": null
29
+ }