Hugging Face's logo

OsaurusAI
/
MiniMax-M2.7-JANGTQ_K

Text Generation
MLX
Safetensors
minimax_m2
osaurus
jangtq
jangtq-prestack
jangtq-k
mixed-precision
minimax
minimax-m2
Mixture of Experts
apple-silicon
conversational
custom_code
Model card Files
xet
 Community
Osaurus-AI commited on
Commit
8716d6a
·
verified ·
1 Parent(s): 552b81e

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +136 -0
config.json ADDED
@@ -0,0 +1,136 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "MiniMaxM2ForCausalLM"
4
+ ],
5
+ "attn_type_list": [
6
+ 1,
7
+ 1,
8
+ 1,
9
+ 1,
10
+ 1,
11
+ 1,
12
+ 1,
13
+ 1,
14
+ 1,
15
+ 1,
16
+ 1,
17
+ 1,
18
+ 1,
19
+ 1,
20
+ 1,
21
+ 1,
22
+ 1,
23
+ 1,
24
+ 1,
25
+ 1,
26
+ 1,
27
+ 1,
28
+ 1,
29
+ 1,
30
+ 1,
31
+ 1,
32
+ 1,
33
+ 1,
34
+ 1,
35
+ 1,
36
+ 1,
37
+ 1,
38
+ 1,
39
+ 1,
40
+ 1,
41
+ 1,
42
+ 1,
43
+ 1,
44
+ 1,
45
+ 1,
46
+ 1,
47
+ 1,
48
+ 1,
49
+ 1,
50
+ 1,
51
+ 1,
52
+ 1,
53
+ 1,
54
+ 1,
55
+ 1,
56
+ 1,
57
+ 1,
58
+ 1,
59
+ 1,
60
+ 1,
61
+ 1,
62
+ 1,
63
+ 1,
64
+ 1,
65
+ 1,
66
+ 1,
67
+ 1
68
+ ],
69
+ "auto_map": {
70
+ "AutoConfig": "configuration_minimax_m2.MiniMaxM2Config",
71
+ "AutoModelForCausalLM": "modeling_minimax_m2.MiniMaxM2ForCausalLM"
72
+ },
73
+ "dtype": "bfloat16",
74
+ "head_dim": 128,
75
+ "hidden_act": "silu",
76
+ "hidden_size": 3072,
77
+ "intermediate_size": 1536,
78
+ "max_position_embeddings": 196608,
79
+ "model_type": "minimax_m2",
80
+ "mtp_transformer_layers": 0,
81
+ "num_attention_heads": 48,
82
+ "num_experts_per_tok": 8,
83
+ "num_hidden_layers": 62,
84
+ "num_key_value_heads": 8,
85
+ "num_local_experts": 256,
86
+ "num_mtp_modules": 0,
87
+ "qk_norm_type": "per_layer",
88
+ "rms_norm_eps": 1e-06,
89
+ "rope_theta": 5000000,
90
+ "rotary_dim": 64,
91
+ "scoring_func": "sigmoid",
92
+ "shared_intermediate_size": 0,
93
+ "tie_word_embeddings": false,
94
+ "transformers_version": "4.46.1",
95
+ "use_cache": true,
96
+ "use_mtp": false,
97
+ "use_qk_norm": true,
98
+ "use_routing_bias": true,
99
+ "vocab_size": 200064,
100
+ "_name_or_path": "M2.5-SLURPY",
101
+ "quantization": {
102
+ "bits": 8,
103
+ "mode": "affine",
104
+ "group_size": 64,
105
+ "routed_expert_bits": {
106
+ "gate_proj": 2,
107
+ "down_proj": 4,
108
+ "up_proj": 2
109
+ },
110
+ "mxtq_bits": {
111
+ "routed_expert": {
112
+ "gate_proj": 2,
113
+ "down_proj": 4,
114
+ "up_proj": 2
115
+ },
116
+ "attention": 8,
117
+ "shared_expert": 8,
118
+ "embed_tokens": 8,
119
+ "lm_head": 8,
120
+ "norms_router_biases": 16
121
+ }
122
+ },
123
+ "mxtq_bits": {
124
+ "routed_expert": {
125
+ "gate_proj": 2,
126
+ "down_proj": 4,
127
+ "up_proj": 2
128
+ },
129
+ "attention": 8,
130
+ "shared_expert": 8,
131
+ "embed_tokens": 8,
132
+ "lm_head": 8,
133
+ "norms_router_biases": 16
134
+ },
135
+ "weight_format": "mxtq"
136
+ }