Hugging Face's logo

FayeQuant
/
GLM-4.7-Flash-GPTQ-4bit

Text Generation
Transformers
Safetensors
English
Chinese
glm4_moe_lite
conversational
4-bit precision
gptq
Model card Files
xet
Community
1
Spica commited on
Commit
6ce1efe
·
verified ·
1 Parent(s): 0b329ba

Upload folder using huggingface_hub

Browse files
Files changed (16) hide show
  1. .gitattributes +1 -0
  2. chat_template.jinja +86 -0
  3. config.json +150 -0
  4. generation_config.json +12 -0
  5. model-00001-of-00008.safetensors +3 -0
  6. model-00002-of-00008.safetensors +3 -0
  7. model-00003-of-00008.safetensors +3 -0
  8. model-00004-of-00008.safetensors +3 -0
  9. model-00005-of-00008.safetensors +3 -0
  10. model-00006-of-00008.safetensors +3 -0
  11. model-00007-of-00008.safetensors +3 -0
  12. model-00008-of-00008.safetensors +3 -0
  13. model.safetensors.index.json +0 -0
  14. quantize_config.json +46 -0
  15. tokenizer.json +3 -0
  16. tokenizer_config.json +12 -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 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [gMASK]<sop>
2
+ {%- if tools -%}
3
+ <|system|>
4
+ # Tools
5
+
6
+ You may call one or more functions to assist with the user query.
7
+
8
+ You are provided with function signatures within <tools></tools> XML tags:
9
+ <tools>
10
+ {% for tool in tools %}
11
+ {{ tool | tojson(ensure_ascii=False) }}
12
+ {% endfor %}
13
+ </tools>
14
+
15
+ For each function call, output the function name and arguments within the following XML format:
16
+ <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 -%}
17
+ {%- macro visible_text(content) -%}
18
+ {%- if content is string -%}
19
+ {{- content }}
20
+ {%- elif content is iterable and content is not mapping -%}
21
+ {%- for item in content -%}
22
+ {%- if item is mapping and item.type == 'text' -%}
23
+ {{- item.text }}
24
+ {%- elif item is string -%}
25
+ {{- item }}
26
+ {%- endif -%}
27
+ {%- endfor -%}
28
+ {%- else -%}
29
+ {{- content }}
30
+ {%- endif -%}
31
+ {%- endmacro -%}
32
+ {%- set ns = namespace(last_user_index=-1) %}
33
+ {%- for m in messages %}
34
+ {%- if m.role == 'user' %}
35
+ {% set ns.last_user_index = loop.index0 -%}
36
+ {%- endif %}
37
+ {%- endfor %}
38
+ {% for m in messages %}
39
+ {%- if m.role == 'user' -%}<|user|>{{ visible_text(m.content) }}
40
+ {%- elif m.role == 'assistant' -%}
41
+ <|assistant|>
42
+ {%- set reasoning_content = '' %}
43
+ {%- set content = visible_text(m.content) %}
44
+ {%- if m.reasoning_content is string %}
45
+ {%- set reasoning_content = m.reasoning_content %}
46
+ {%- else %}
47
+ {%- if '</think>' in content %}
48
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
49
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
50
+ {%- endif %}
51
+ {%- endif %}
52
+ {%- if ((clear_thinking is defined and not clear_thinking) or loop.index0 > ns.last_user_index) and reasoning_content -%}
53
+ {{ '<think>' + reasoning_content.strip() + '</think>'}}
54
+ {%- else -%}
55
+ {{ '</think>' }}
56
+ {%- endif -%}
57
+ {%- if content.strip() -%}
58
+ {{ content.strip() }}
59
+ {%- endif -%}
60
+ {% if m.tool_calls %}
61
+ {% for tc in m.tool_calls %}
62
+ {%- if tc.function %}
63
+ {%- set tc = tc.function %}
64
+ {%- endif %}
65
+ {{- '<tool_call>' + tc.name -}}
66
+ {% 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 %}
67
+ {% endif %}
68
+ {%- elif m.role == 'tool' -%}
69
+ {%- if m.content is string -%}
70
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
71
+ {{- '<|observation|>' }}
72
+ {%- endif %}
73
+ {{- '<tool_response>' }}
74
+ {{- m.content }}
75
+ {{- '</tool_response>' }}
76
+ {%- else -%}
77
+ <|observation|>{% for tr in m.content %}
78
+ <tool_response>{{ tr.output if tr.output is defined else tr }}</tool_response>{% endfor -%}
79
+ {% endif -%}
80
+ {%- elif m.role == 'system' -%}
81
+ <|system|>{{ visible_text(m.content) }}
82
+ {%- endif -%}
83
+ {%- endfor -%}
84
+ {%- if add_generation_prompt -%}
85
+ <|assistant|>{{- '</think>' if (enable_thinking is defined and not enable_thinking) else '<think>' -}}
86
+ {%- endif -%}
config.json ADDED
@@ -0,0 +1,150 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Glm4MoeLiteForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 0,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": [
10
+ 154820,
11
+ 154827,
12
+ 154829
13
+ ],
14
+ "first_k_dense_replace": 1,
15
+ "head_dim": 64,
16
+ "hidden_act": "silu",
17
+ "hidden_size": 2048,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 10240,
20
+ "kv_lora_rank": 512,
21
+ "max_position_embeddings": 202752,
22
+ "mlp_layer_types": [
23
+ "dense",
24
+ "sparse",
25
+ "sparse",
26
+ "sparse",
27
+ "sparse",
28
+ "sparse",
29
+ "sparse",
30
+ "sparse",
31
+ "sparse",
32
+ "sparse",
33
+ "sparse",
34
+ "sparse",
35
+ "sparse",
36
+ "sparse",
37
+ "sparse",
38
+ "sparse",
39
+ "sparse",
40
+ "sparse",
41
+ "sparse",
42
+ "sparse",
43
+ "sparse",
44
+ "sparse",
45
+ "sparse",
46
+ "sparse",
47
+ "sparse",
48
+ "sparse",
49
+ "sparse",
50
+ "sparse",
51
+ "sparse",
52
+ "sparse",
53
+ "sparse",
54
+ "sparse",
55
+ "sparse",
56
+ "sparse",
57
+ "sparse",
58
+ "sparse",
59
+ "sparse",
60
+ "sparse",
61
+ "sparse",
62
+ "sparse",
63
+ "sparse",
64
+ "sparse",
65
+ "sparse",
66
+ "sparse",
67
+ "sparse",
68
+ "sparse",
69
+ "sparse"
70
+ ],
71
+ "model_type": "glm4_moe_lite",
72
+ "moe_intermediate_size": 1536,
73
+ "n_group": 1,
74
+ "n_routed_experts": 64,
75
+ "n_shared_experts": 1,
76
+ "norm_topk_prob": true,
77
+ "num_attention_heads": 20,
78
+ "num_experts_per_tok": 4,
79
+ "num_hidden_layers": 47,
80
+ "num_key_value_heads": 20,
81
+ "num_nextn_predict_layers": 1,
82
+ "pad_token_id": 154820,
83
+ "partial_rotary_factor": 1.0,
84
+ "pretraining_tp": 1,
85
+ "q_lora_rank": 768,
86
+ "qk_head_dim": 256,
87
+ "qk_nope_head_dim": 192,
88
+ "qk_rope_head_dim": 64,
89
+ "quantization_config": {
90
+ "bits": 4,
91
+ "checkpoint_format": "gptq",
92
+ "desc_act": false,
93
+ "format": "gptq",
94
+ "group_size": 128,
95
+ "lm_head": false,
96
+ "meta": {
97
+ "act_group_aware": true,
98
+ "auto_forward_data_parallel": true,
99
+ "damp_auto_increment": 0.01,
100
+ "damp_percent": 0.05,
101
+ "failsafe": {
102
+ "smooth": {
103
+ "group_size_threshold": 128,
104
+ "k": 2.75,
105
+ "type": "mad"
106
+ },
107
+ "strategy": "rtn",
108
+ "threshold": "0.5%"
109
+ },
110
+ "gc_mode": "interval",
111
+ "gptaq": null,
112
+ "hessian": {
113
+ "chunk_bytes": null,
114
+ "chunk_size": null,
115
+ "staging_dtype": "float32"
116
+ },
117
+ "mock_quantization": false,
118
+ "mse": 0.0,
119
+ "offload_to_disk": true,
120
+ "offload_to_disk_path": "./gptqmodel_offload/hqdpgrum-rkaakpxx/",
121
+ "pack_impl": "cpu",
122
+ "quantizer": [
123
+ "gptqmodel:5.7.0"
124
+ ],
125
+ "static_groups": false,
126
+ "true_sequential": true,
127
+ "uri": "https://github.com/modelcloud/gptqmodel",
128
+ "vram_strategy": "exclusive",
129
+ "wait_for_submodule_finalizers": false
130
+ },
131
+ "pack_dtype": "int32",
132
+ "quant_method": "gptq",
133
+ "sym": true
134
+ },
135
+ "rms_norm_eps": 1e-05,
136
+ "rope_interleave": true,
137
+ "rope_parameters": {
138
+ "partial_rotary_factor": 1.0,
139
+ "rope_theta": 1000000,
140
+ "rope_type": "default"
141
+ },
142
+ "routed_scaling_factor": 1.8,
143
+ "tie_word_embeddings": false,
144
+ "topk_group": 1,
145
+ "topk_method": "noaux_tc",
146
+ "transformers_version": "5.1.0",
147
+ "use_cache": true,
148
+ "v_head_dim": 256,
149
+ "vocab_size": 154880
150
+ }
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 154820,
6
+ 154827,
7
+ 154829
8
+ ],
9
+ "pad_token_id": 154820,
10
+ "temperature": 1.0,
11
+ "transformers_version": "5.1.0"
12
+ }
model-00001-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08468c358b6008ff8e24c3781d1f1f71bc80c23e86e28d3289afb8f348f66207
3
+ size 2146226724
model-00002-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d56ec0a2412cb8820f4c27d75638293e7d08c124b0c3768162ae1aa2598ab39
3
+ size 2147232287
model-00003-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:29c5e627f09e2974806a6d61320b7493a102e1f0c7a4231dc31e3d5ebc91283a
3
+ size 2147091286
model-00004-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cce672a91ba02a79853ee167abf4ea8fda577903e978a6e8b52d0eb320f6640f
3
+ size 2147237477
model-00005-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b4abd89d54f58e3129dcaca8aa87bf972654b007801909f86460c7e86e0c101d
3
+ size 2147089277
model-00006-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5c19bf4264c5a133fc02c06253c1e13b40b2425a7aac89b795e3b84d7c29f6b
3
+ size 2147237477
model-00007-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67737a42c82379b68464a23f357d116b7a2e9cc3fa84918c359052be721cdbda
3
+ size 2147089273
model-00008-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa27dd7ee7527bccbc384f4e5eeb288635f042a1451d82338ab814440ff4a3d8
3
+ size 1549678974
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
quantize_config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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:5.7.0"
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": {
25
+ "type": "mad",
26
+ "group_size_threshold": 128,
27
+ "k": 2.75
28
+ }
29
+ },
30
+ "offload_to_disk": true,
31
+ "offload_to_disk_path": "./gptqmodel_offload/hqdpgrum-rkaakpxx/",
32
+ "pack_impl": "cpu",
33
+ "mock_quantization": false,
34
+ "gc_mode": "interval",
35
+ "wait_for_submodule_finalizers": false,
36
+ "auto_forward_data_parallel": true,
37
+ "hessian": {
38
+ "chunk_size": null,
39
+ "chunk_bytes": null,
40
+ "staging_dtype": "float32"
41
+ },
42
+ "vram_strategy": "exclusive"
43
+ },
44
+ "sym": true,
45
+ "format": "gptq"
46
+ }
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 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "clean_up_tokenization_spaces": false,
4
+ "do_lower_case": false,
5
+ "eos_token": "<|endoftext|>",
6
+ "is_local": false,
7
+ "model_max_length": 128000,
8
+ "pad_token": "<|endoftext|>",
9
+ "padding_side": "left",
10
+ "remove_space": false,
11
+ "tokenizer_class": "TokenizersBackend"
12
+ }