Hugging Face's logo

mattbucci
/
Qwen3-Coder-Next-REAM-AWQ

Safetensors
qwen3_next
4-bit precision
awq
Model card Files
xet
 Community
mattbucci commited on
Commit
a048d7b
·
verified ·
1 Parent(s): 8ab3494

add config.json

Browse files
Files changed (1) hide show
  1. config.json +129 -0
config.json ADDED
@@ -0,0 +1,129 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3NextForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0,
7
+ "bos_token_id": 151643,
8
+ "decoder_sparse_step": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 151645,
11
+ "full_attention_interval": 4,
12
+ "head_dim": 256,
13
+ "hidden_act": "silu",
14
+ "hidden_size": 2048,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 5120,
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
+ "linear_attention",
51
+ "linear_attention",
52
+ "linear_attention",
53
+ "full_attention",
54
+ "linear_attention",
55
+ "linear_attention",
56
+ "linear_attention",
57
+ "full_attention",
58
+ "linear_attention",
59
+ "linear_attention",
60
+ "linear_attention",
61
+ "full_attention",
62
+ "linear_attention",
63
+ "linear_attention",
64
+ "linear_attention",
65
+ "full_attention"
66
+ ],
67
+ "linear_conv_kernel_dim": 4,
68
+ "linear_key_head_dim": 128,
69
+ "linear_num_key_heads": 16,
70
+ "linear_num_value_heads": 32,
71
+ "linear_value_head_dim": 128,
72
+ "max_position_embeddings": 262144,
73
+ "merge_args": {
74
+ "balance_group_size": 32,
75
+ "dataset": "c4+math+the-stack-smol",
76
+ "expert_saliency": "reap",
77
+ "gate_softmax": true,
78
+ "group": "freq_logits",
79
+ "merge": "align_logits_weights",
80
+ "merge_size": 384,
81
+ "merger_bs": 3072,
82
+ "merger_seq_len": 512,
83
+ "pca_dim": 64,
84
+ "precompute_input": false,
85
+ "use_gate_output": true
86
+ },
87
+ "mlp_only_layers": [],
88
+ "model_type": "qwen3_next",
89
+ "moe_intermediate_size": 512,
90
+ "norm_topk_prob": true,
91
+ "num_attention_heads": 16,
92
+ "num_experts": 384,
93
+ "num_experts_per_tok": 10,
94
+ "num_hidden_layers": 48,
95
+ "num_key_value_heads": 2,
96
+ "output_router_logits": false,
97
+ "partial_rotary_factor": 0.25,
98
+ "quantization_config": {
99
+ "bits": 4,
100
+ "group_size": 32,
101
+ "quant_method": "awq",
102
+ "version": "gemm",
103
+ "zero_point": true,
104
+ "modules_to_not_convert": [
105
+ "lm_head",
106
+ "linear_attn",
107
+ "self_attn",
108
+ "shared_expert",
109
+ "shared_expert_gate",
110
+ "gate",
111
+ "input_layernorm",
112
+ "post_attention_layernorm",
113
+ "embed_tokens",
114
+ "norm",
115
+ "k_norm",
116
+ "q_norm"
117
+ ]
118
+ },
119
+ "rms_norm_eps": 1e-06,
120
+ "rope_scaling": null,
121
+ "rope_theta": 5000000,
122
+ "router_aux_loss_coef": 0.001,
123
+ "shared_expert_intermediate_size": 512,
124
+ "tie_word_embeddings": false,
125
+ "transformers_version": "4.57.3",
126
+ "use_cache": true,
127
+ "use_sliding_window": false,
128
+ "vocab_size": 151936
129
+ }