jsbai-aaron commited on
Commit
62caa4d
·
verified ·
1 Parent(s): 10b2e22

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +146 -0
config.json ADDED
@@ -0,0 +1,146 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3_5ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attn_output_gate": true,
8
+ "bos_token_id": null,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 248044,
11
+ "full_attention_interval": 4,
12
+ "head_dim": 256,
13
+ "hidden_act": "silu",
14
+ "hidden_size": 2560,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 9216,
17
+ "layer_types": [
18
+ "linear_attention",
19
+ "linear_attention",
20
+ "linear_attention",
21
+ "full_attention",
22
+ "linear_attention",
23
+ "linear_attention",
24
+ "linear_attention",
25
+ "full_attention",
26
+ "linear_attention",
27
+ "linear_attention",
28
+ "linear_attention",
29
+ "full_attention",
30
+ "linear_attention",
31
+ "linear_attention",
32
+ "linear_attention",
33
+ "full_attention",
34
+ "linear_attention",
35
+ "linear_attention",
36
+ "linear_attention",
37
+ "full_attention",
38
+ "linear_attention",
39
+ "linear_attention",
40
+ "linear_attention",
41
+ "full_attention",
42
+ "linear_attention",
43
+ "linear_attention",
44
+ "linear_attention",
45
+ "full_attention",
46
+ "linear_attention",
47
+ "linear_attention",
48
+ "linear_attention",
49
+ "full_attention"
50
+ ],
51
+ "linear_conv_kernel_dim": 4,
52
+ "linear_key_head_dim": 128,
53
+ "linear_num_key_heads": 16,
54
+ "linear_num_value_heads": 32,
55
+ "linear_value_head_dim": 128,
56
+ "mamba_ssm_dtype": "float32",
57
+ "max_position_embeddings": 262144,
58
+ "mlp_only_layers": [],
59
+ "model_type": "qwen3_5_text",
60
+ "mtp_num_hidden_layers": 1,
61
+ "mtp_use_dedicated_embeddings": false,
62
+ "num_attention_heads": 16,
63
+ "num_hidden_layers": 32,
64
+ "num_key_value_heads": 4,
65
+ "pad_token_id": null,
66
+ "partial_rotary_factor": 0.25,
67
+ "rms_norm_eps": 1e-06,
68
+ "rope_parameters": {
69
+ "mrope_interleaved": true,
70
+ "mrope_section": [
71
+ 11,
72
+ 11,
73
+ 10
74
+ ],
75
+ "partial_rotary_factor": 0.25,
76
+ "rope_theta": 10000000,
77
+ "rope_type": "default"
78
+ },
79
+ "tie_word_embeddings": true,
80
+ "transformers_version": "5.16.1",
81
+ "use_cache": true,
82
+ "use_mamba_kernels": true,
83
+ "vocab_size": 248320,
84
+ "quantization_config": {
85
+ "config_groups": {
86
+ "group_0": {
87
+ "input_activations": {
88
+ "dynamic": false,
89
+ "num_bits": 4,
90
+ "type": "float",
91
+ "group_size": 16
92
+ },
93
+ "weights": {
94
+ "dynamic": false,
95
+ "num_bits": 4,
96
+ "type": "float",
97
+ "group_size": 16
98
+ },
99
+ "targets": [
100
+ "Linear"
101
+ ]
102
+ }
103
+ },
104
+ "ignore": [
105
+ "lm_head",
106
+ "model.layers.0.linear_attn*",
107
+ "model.layers.1.linear_attn*",
108
+ "model.layers.10.linear_attn*",
109
+ "model.layers.11.self_attn*",
110
+ "model.layers.12.linear_attn*",
111
+ "model.layers.13.linear_attn*",
112
+ "model.layers.14.linear_attn*",
113
+ "model.layers.15.self_attn*",
114
+ "model.layers.16.linear_attn*",
115
+ "model.layers.17.linear_attn*",
116
+ "model.layers.18.linear_attn*",
117
+ "model.layers.19.self_attn*",
118
+ "model.layers.2.linear_attn*",
119
+ "model.layers.20.linear_attn*",
120
+ "model.layers.21.linear_attn*",
121
+ "model.layers.22.linear_attn*",
122
+ "model.layers.23.self_attn*",
123
+ "model.layers.24.linear_attn*",
124
+ "model.layers.25.linear_attn*",
125
+ "model.layers.26.linear_attn*",
126
+ "model.layers.27.self_attn*",
127
+ "model.layers.28.linear_attn*",
128
+ "model.layers.29.linear_attn*",
129
+ "model.layers.3.self_attn*",
130
+ "model.layers.30.linear_attn*",
131
+ "model.layers.31.self_attn*",
132
+ "model.layers.4.linear_attn*",
133
+ "model.layers.5.linear_attn*",
134
+ "model.layers.6.linear_attn*",
135
+ "model.layers.7.self_attn*",
136
+ "model.layers.8.linear_attn*",
137
+ "model.layers.9.linear_attn*"
138
+ ],
139
+ "quant_algo": "NVFP4",
140
+ "producer": {
141
+ "name": "modelopt",
142
+ "version": "0.44.0"
143
+ },
144
+ "quant_method": "modelopt"
145
+ }
146
+ }