Hugging Face's logo

smcleod
/
dots.tts-soar-mlx

Text-to-Speech
MLX
Safetensors
English
dots_tts
tts
quantized
4-bit precision
8-bit precision
apple-silicon
dots.tts
Model card Files
xet
 Community
smcleod commited on
Commit
16fa129
·
verified ·
1 Parent(s): 9005175

Add MeanFlow NFE=4 variant (mf/): int8 meanflow DiT + shared 8bit components

Browse files
Files changed (20) hide show
  1. .gitattributes +1 -0
  2. mf/audiovae_encoder/model.safetensors +3 -0
  3. mf/backbone/README.md +7 -0
  4. mf/backbone/config.json +38 -0
  5. mf/backbone/model.safetensors +3 -0
  6. mf/backbone/model.safetensors.index.json +740 -0
  7. mf/backbone/tokenizer.json +3 -0
  8. mf/backbone/tokenizer_config.json +15 -0
  9. mf/config.json +113 -0
  10. mf/dit/config.json +22 -0
  11. mf/dit/model.safetensors +3 -0
  12. mf/heads/model.safetensors +3 -0
  13. mf/latent_stats.json +1 -0
  14. mf/latent_stats.pt +3 -0
  15. mf/llm_config.json +114 -0
  16. mf/patch_encoder/config.json +7 -0
  17. mf/patch_encoder/model.safetensors +3 -0
  18. mf/resampler_48k_16k.safetensors +3 -0
  19. mf/speaker/model.safetensors +3 -0
  20. mf/vocoder/model.safetensors +3 -0
.gitattributes CHANGED
@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
36
  4bit/backbone/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
  8bit/backbone/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
  backbone/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
36
  4bit/backbone/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
  8bit/backbone/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
  backbone/tokenizer.json filter=lfs diff=lfs merge=lfs -text
39
+ mf/backbone/tokenizer.json filter=lfs diff=lfs merge=lfs -text
mf/audiovae_encoder/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8091838d924f821ec38968aa922e3d9325c45abb88e1033ddb72c44827a17dc3
3
+ size 88700706
mf/backbone/README.md ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ pipeline_tag: text-generation
4
+ library_name: mlx
5
+ tags:
6
+ - mlx
7
+ ---
mf/backbone/config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen2ForCausalLM"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "bos_token_id": 151643,
7
+ "eos_token_id": 151643,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 1536,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 8960,
12
+ "max_position_embeddings": 131072,
13
+ "max_window_layers": 28,
14
+ "model_type": "qwen2",
15
+ "num_attention_heads": 12,
16
+ "num_hidden_layers": 28,
17
+ "num_key_value_heads": 2,
18
+ "quantization": {
19
+ "group_size": 64,
20
+ "bits": 8,
21
+ "mode": "affine"
22
+ },
23
+ "quantization_config": {
24
+ "group_size": 64,
25
+ "bits": 8,
26
+ "mode": "affine"
27
+ },
28
+ "rms_norm_eps": 1e-06,
29
+ "rope_scaling": null,
30
+ "rope_theta": 1000000.0,
31
+ "sliding_window": null,
32
+ "tie_word_embeddings": true,
33
+ "torch_dtype": "bfloat16",
34
+ "transformers_version": "4.57.0",
35
+ "use_cache": true,
36
+ "use_sliding_window": false,
37
+ "vocab_size": 151672
38
+ }
mf/backbone/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1312c10250b3b304da1e5e00a63850c46e1b153f999a7930eaa61d6e139375a
3
+ size 1639983513
mf/backbone/model.safetensors.index.json ADDED
@@ -0,0 +1,740 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 1639901440,
4
+ "total_parameters": 1543308800
5
+ },
6
+ "weight_map": {
7
+ "model.embed_tokens.biases": "model.safetensors",
8
+ "model.embed_tokens.scales": "model.safetensors",
9
+ "model.embed_tokens.weight": "model.safetensors",
10
+ "model.layers.0.input_layernorm.weight": "model.safetensors",
11
+ "model.layers.0.mlp.down_proj.biases": "model.safetensors",
12
+ "model.layers.0.mlp.down_proj.scales": "model.safetensors",
13
+ "model.layers.0.mlp.down_proj.weight": "model.safetensors",
14
+ "model.layers.0.mlp.gate_proj.biases": "model.safetensors",
15
+ "model.layers.0.mlp.gate_proj.scales": "model.safetensors",
16
+ "model.layers.0.mlp.gate_proj.weight": "model.safetensors",
17
+ "model.layers.0.mlp.up_proj.biases": "model.safetensors",
18
+ "model.layers.0.mlp.up_proj.scales": "model.safetensors",
19
+ "model.layers.0.mlp.up_proj.weight": "model.safetensors",
20
+ "model.layers.0.post_attention_layernorm.weight": "model.safetensors",
21
+ "model.layers.0.self_attn.k_proj.bias": "model.safetensors",
22
+ "model.layers.0.self_attn.k_proj.biases": "model.safetensors",
23
+ "model.layers.0.self_attn.k_proj.scales": "model.safetensors",
24
+ "model.layers.0.self_attn.k_proj.weight": "model.safetensors",
25
+ "model.layers.0.self_attn.o_proj.biases": "model.safetensors",
26
+ "model.layers.0.self_attn.o_proj.scales": "model.safetensors",
27
+ "model.layers.0.self_attn.o_proj.weight": "model.safetensors",
28
+ "model.layers.0.self_attn.q_proj.bias": "model.safetensors",
29
+ "model.layers.0.self_attn.q_proj.biases": "model.safetensors",
30
+ "model.layers.0.self_attn.q_proj.scales": "model.safetensors",
31
+ "model.layers.0.self_attn.q_proj.weight": "model.safetensors",
32
+ "model.layers.0.self_attn.v_proj.bias": "model.safetensors",
33
+ "model.layers.0.self_attn.v_proj.biases": "model.safetensors",
34
+ "model.layers.0.self_attn.v_proj.scales": "model.safetensors",
35
+ "model.layers.0.self_attn.v_proj.weight": "model.safetensors",
36
+ "model.layers.1.input_layernorm.weight": "model.safetensors",
37
+ "model.layers.1.mlp.down_proj.biases": "model.safetensors",
38
+ "model.layers.1.mlp.down_proj.scales": "model.safetensors",
39
+ "model.layers.1.mlp.down_proj.weight": "model.safetensors",
40
+ "model.layers.1.mlp.gate_proj.biases": "model.safetensors",
41
+ "model.layers.1.mlp.gate_proj.scales": "model.safetensors",
42
+ "model.layers.1.mlp.gate_proj.weight": "model.safetensors",
43
+ "model.layers.1.mlp.up_proj.biases": "model.safetensors",
44
+ "model.layers.1.mlp.up_proj.scales": "model.safetensors",
45
+ "model.layers.1.mlp.up_proj.weight": "model.safetensors",
46
+ "model.layers.1.post_attention_layernorm.weight": "model.safetensors",
47
+ "model.layers.1.self_attn.k_proj.bias": "model.safetensors",
48
+ "model.layers.1.self_attn.k_proj.biases": "model.safetensors",
49
+ "model.layers.1.self_attn.k_proj.scales": "model.safetensors",
50
+ "model.layers.1.self_attn.k_proj.weight": "model.safetensors",
51
+ "model.layers.1.self_attn.o_proj.biases": "model.safetensors",
52
+ "model.layers.1.self_attn.o_proj.scales": "model.safetensors",
53
+ "model.layers.1.self_attn.o_proj.weight": "model.safetensors",
54
+ "model.layers.1.self_attn.q_proj.bias": "model.safetensors",
55
+ "model.layers.1.self_attn.q_proj.biases": "model.safetensors",
56
+ "model.layers.1.self_attn.q_proj.scales": "model.safetensors",
57
+ "model.layers.1.self_attn.q_proj.weight": "model.safetensors",
58
+ "model.layers.1.self_attn.v_proj.bias": "model.safetensors",
59
+ "model.layers.1.self_attn.v_proj.biases": "model.safetensors",
60
+ "model.layers.1.self_attn.v_proj.scales": "model.safetensors",
61
+ "model.layers.1.self_attn.v_proj.weight": "model.safetensors",
62
+ "model.layers.10.input_layernorm.weight": "model.safetensors",
63
+ "model.layers.10.mlp.down_proj.biases": "model.safetensors",
64
+ "model.layers.10.mlp.down_proj.scales": "model.safetensors",
65
+ "model.layers.10.mlp.down_proj.weight": "model.safetensors",
66
+ "model.layers.10.mlp.gate_proj.biases": "model.safetensors",
67
+ "model.layers.10.mlp.gate_proj.scales": "model.safetensors",
68
+ "model.layers.10.mlp.gate_proj.weight": "model.safetensors",
69
+ "model.layers.10.mlp.up_proj.biases": "model.safetensors",
70
+ "model.layers.10.mlp.up_proj.scales": "model.safetensors",
71
+ "model.layers.10.mlp.up_proj.weight": "model.safetensors",
72
+ "model.layers.10.post_attention_layernorm.weight": "model.safetensors",
73
+ "model.layers.10.self_attn.k_proj.bias": "model.safetensors",
74
+ "model.layers.10.self_attn.k_proj.biases": "model.safetensors",
75
+ "model.layers.10.self_attn.k_proj.scales": "model.safetensors",
76
+ "model.layers.10.self_attn.k_proj.weight": "model.safetensors",
77
+ "model.layers.10.self_attn.o_proj.biases": "model.safetensors",
78
+ "model.layers.10.self_attn.o_proj.scales": "model.safetensors",
79
+ "model.layers.10.self_attn.o_proj.weight": "model.safetensors",
80
+ "model.layers.10.self_attn.q_proj.bias": "model.safetensors",
81
+ "model.layers.10.self_attn.q_proj.biases": "model.safetensors",
82
+ "model.layers.10.self_attn.q_proj.scales": "model.safetensors",
83
+ "model.layers.10.self_attn.q_proj.weight": "model.safetensors",
84
+ "model.layers.10.self_attn.v_proj.bias": "model.safetensors",
85
+ "model.layers.10.self_attn.v_proj.biases": "model.safetensors",
86
+ "model.layers.10.self_attn.v_proj.scales": "model.safetensors",
87
+ "model.layers.10.self_attn.v_proj.weight": "model.safetensors",
88
+ "model.layers.11.input_layernorm.weight": "model.safetensors",
89
+ "model.layers.11.mlp.down_proj.biases": "model.safetensors",
90
+ "model.layers.11.mlp.down_proj.scales": "model.safetensors",
91
+ "model.layers.11.mlp.down_proj.weight": "model.safetensors",
92
+ "model.layers.11.mlp.gate_proj.biases": "model.safetensors",
93
+ "model.layers.11.mlp.gate_proj.scales": "model.safetensors",
94
+ "model.layers.11.mlp.gate_proj.weight": "model.safetensors",
95
+ "model.layers.11.mlp.up_proj.biases": "model.safetensors",
96
+ "model.layers.11.mlp.up_proj.scales": "model.safetensors",
97
+ "model.layers.11.mlp.up_proj.weight": "model.safetensors",
98
+ "model.layers.11.post_attention_layernorm.weight": "model.safetensors",
99
+ "model.layers.11.self_attn.k_proj.bias": "model.safetensors",
100
+ "model.layers.11.self_attn.k_proj.biases": "model.safetensors",
101
+ "model.layers.11.self_attn.k_proj.scales": "model.safetensors",
102
+ "model.layers.11.self_attn.k_proj.weight": "model.safetensors",
103
+ "model.layers.11.self_attn.o_proj.biases": "model.safetensors",
104
+ "model.layers.11.self_attn.o_proj.scales": "model.safetensors",
105
+ "model.layers.11.self_attn.o_proj.weight": "model.safetensors",
106
+ "model.layers.11.self_attn.q_proj.bias": "model.safetensors",
107
+ "model.layers.11.self_attn.q_proj.biases": "model.safetensors",
108
+ "model.layers.11.self_attn.q_proj.scales": "model.safetensors",
109
+ "model.layers.11.self_attn.q_proj.weight": "model.safetensors",
110
+ "model.layers.11.self_attn.v_proj.bias": "model.safetensors",
111
+ "model.layers.11.self_attn.v_proj.biases": "model.safetensors",
112
+ "model.layers.11.self_attn.v_proj.scales": "model.safetensors",
113
+ "model.layers.11.self_attn.v_proj.weight": "model.safetensors",
114
+ "model.layers.12.input_layernorm.weight": "model.safetensors",
115
+ "model.layers.12.mlp.down_proj.biases": "model.safetensors",
116
+ "model.layers.12.mlp.down_proj.scales": "model.safetensors",
117
+ "model.layers.12.mlp.down_proj.weight": "model.safetensors",
118
+ "model.layers.12.mlp.gate_proj.biases": "model.safetensors",
119
+ "model.layers.12.mlp.gate_proj.scales": "model.safetensors",
120
+ "model.layers.12.mlp.gate_proj.weight": "model.safetensors",
121
+ "model.layers.12.mlp.up_proj.biases": "model.safetensors",
122
+ "model.layers.12.mlp.up_proj.scales": "model.safetensors",
123
+ "model.layers.12.mlp.up_proj.weight": "model.safetensors",
124
+ "model.layers.12.post_attention_layernorm.weight": "model.safetensors",
125
+ "model.layers.12.self_attn.k_proj.bias": "model.safetensors",
126
+ "model.layers.12.self_attn.k_proj.biases": "model.safetensors",
127
+ "model.layers.12.self_attn.k_proj.scales": "model.safetensors",
128
+ "model.layers.12.self_attn.k_proj.weight": "model.safetensors",
129
+ "model.layers.12.self_attn.o_proj.biases": "model.safetensors",
130
+ "model.layers.12.self_attn.o_proj.scales": "model.safetensors",
131
+ "model.layers.12.self_attn.o_proj.weight": "model.safetensors",
132
+ "model.layers.12.self_attn.q_proj.bias": "model.safetensors",
133
+ "model.layers.12.self_attn.q_proj.biases": "model.safetensors",
134
+ "model.layers.12.self_attn.q_proj.scales": "model.safetensors",
135
+ "model.layers.12.self_attn.q_proj.weight": "model.safetensors",
136
+ "model.layers.12.self_attn.v_proj.bias": "model.safetensors",
137
+ "model.layers.12.self_attn.v_proj.biases": "model.safetensors",
138
+ "model.layers.12.self_attn.v_proj.scales": "model.safetensors",
139
+ "model.layers.12.self_attn.v_proj.weight": "model.safetensors",
140
+ "model.layers.13.input_layernorm.weight": "model.safetensors",
141
+ "model.layers.13.mlp.down_proj.biases": "model.safetensors",
142
+ "model.layers.13.mlp.down_proj.scales": "model.safetensors",
143
+ "model.layers.13.mlp.down_proj.weight": "model.safetensors",
144
+ "model.layers.13.mlp.gate_proj.biases": "model.safetensors",
145
+ "model.layers.13.mlp.gate_proj.scales": "model.safetensors",
146
+ "model.layers.13.mlp.gate_proj.weight": "model.safetensors",
147
+ "model.layers.13.mlp.up_proj.biases": "model.safetensors",
148
+ "model.layers.13.mlp.up_proj.scales": "model.safetensors",
149
+ "model.layers.13.mlp.up_proj.weight": "model.safetensors",
150
+ "model.layers.13.post_attention_layernorm.weight": "model.safetensors",
151
+ "model.layers.13.self_attn.k_proj.bias": "model.safetensors",
152
+ "model.layers.13.self_attn.k_proj.biases": "model.safetensors",
153
+ "model.layers.13.self_attn.k_proj.scales": "model.safetensors",
154
+ "model.layers.13.self_attn.k_proj.weight": "model.safetensors",
155
+ "model.layers.13.self_attn.o_proj.biases": "model.safetensors",
156
+ "model.layers.13.self_attn.o_proj.scales": "model.safetensors",
157
+ "model.layers.13.self_attn.o_proj.weight": "model.safetensors",
158
+ "model.layers.13.self_attn.q_proj.bias": "model.safetensors",
159
+ "model.layers.13.self_attn.q_proj.biases": "model.safetensors",
160
+ "model.layers.13.self_attn.q_proj.scales": "model.safetensors",
161
+ "model.layers.13.self_attn.q_proj.weight": "model.safetensors",
162
+ "model.layers.13.self_attn.v_proj.bias": "model.safetensors",
163
+ "model.layers.13.self_attn.v_proj.biases": "model.safetensors",
164
+ "model.layers.13.self_attn.v_proj.scales": "model.safetensors",
165
+ "model.layers.13.self_attn.v_proj.weight": "model.safetensors",
166
+ "model.layers.14.input_layernorm.weight": "model.safetensors",
167
+ "model.layers.14.mlp.down_proj.biases": "model.safetensors",
168
+ "model.layers.14.mlp.down_proj.scales": "model.safetensors",
169
+ "model.layers.14.mlp.down_proj.weight": "model.safetensors",
170
+ "model.layers.14.mlp.gate_proj.biases": "model.safetensors",
171
+ "model.layers.14.mlp.gate_proj.scales": "model.safetensors",
172
+ "model.layers.14.mlp.gate_proj.weight": "model.safetensors",
173
+ "model.layers.14.mlp.up_proj.biases": "model.safetensors",
174
+ "model.layers.14.mlp.up_proj.scales": "model.safetensors",
175
+ "model.layers.14.mlp.up_proj.weight": "model.safetensors",
176
+ "model.layers.14.post_attention_layernorm.weight": "model.safetensors",
177
+ "model.layers.14.self_attn.k_proj.bias": "model.safetensors",
178
+ "model.layers.14.self_attn.k_proj.biases": "model.safetensors",
179
+ "model.layers.14.self_attn.k_proj.scales": "model.safetensors",
180
+ "model.layers.14.self_attn.k_proj.weight": "model.safetensors",
181
+ "model.layers.14.self_attn.o_proj.biases": "model.safetensors",
182
+ "model.layers.14.self_attn.o_proj.scales": "model.safetensors",
183
+ "model.layers.14.self_attn.o_proj.weight": "model.safetensors",
184
+ "model.layers.14.self_attn.q_proj.bias": "model.safetensors",
185
+ "model.layers.14.self_attn.q_proj.biases": "model.safetensors",
186
+ "model.layers.14.self_attn.q_proj.scales": "model.safetensors",
187
+ "model.layers.14.self_attn.q_proj.weight": "model.safetensors",
188
+ "model.layers.14.self_attn.v_proj.bias": "model.safetensors",
189
+ "model.layers.14.self_attn.v_proj.biases": "model.safetensors",
190
+ "model.layers.14.self_attn.v_proj.scales": "model.safetensors",
191
+ "model.layers.14.self_attn.v_proj.weight": "model.safetensors",
192
+ "model.layers.15.input_layernorm.weight": "model.safetensors",
193
+ "model.layers.15.mlp.down_proj.biases": "model.safetensors",
194
+ "model.layers.15.mlp.down_proj.scales": "model.safetensors",
195
+ "model.layers.15.mlp.down_proj.weight": "model.safetensors",
196
+ "model.layers.15.mlp.gate_proj.biases": "model.safetensors",
197
+ "model.layers.15.mlp.gate_proj.scales": "model.safetensors",
198
+ "model.layers.15.mlp.gate_proj.weight": "model.safetensors",
199
+ "model.layers.15.mlp.up_proj.biases": "model.safetensors",
200
+ "model.layers.15.mlp.up_proj.scales": "model.safetensors",
201
+ "model.layers.15.mlp.up_proj.weight": "model.safetensors",
202
+ "model.layers.15.post_attention_layernorm.weight": "model.safetensors",
203
+ "model.layers.15.self_attn.k_proj.bias": "model.safetensors",
204
+ "model.layers.15.self_attn.k_proj.biases": "model.safetensors",
205
+ "model.layers.15.self_attn.k_proj.scales": "model.safetensors",
206
+ "model.layers.15.self_attn.k_proj.weight": "model.safetensors",
207
+ "model.layers.15.self_attn.o_proj.biases": "model.safetensors",
208
+ "model.layers.15.self_attn.o_proj.scales": "model.safetensors",
209
+ "model.layers.15.self_attn.o_proj.weight": "model.safetensors",
210
+ "model.layers.15.self_attn.q_proj.bias": "model.safetensors",
211
+ "model.layers.15.self_attn.q_proj.biases": "model.safetensors",
212
+ "model.layers.15.self_attn.q_proj.scales": "model.safetensors",
213
+ "model.layers.15.self_attn.q_proj.weight": "model.safetensors",
214
+ "model.layers.15.self_attn.v_proj.bias": "model.safetensors",
215
+ "model.layers.15.self_attn.v_proj.biases": "model.safetensors",
216
+ "model.layers.15.self_attn.v_proj.scales": "model.safetensors",
217
+ "model.layers.15.self_attn.v_proj.weight": "model.safetensors",
218
+ "model.layers.16.input_layernorm.weight": "model.safetensors",
219
+ "model.layers.16.mlp.down_proj.biases": "model.safetensors",
220
+ "model.layers.16.mlp.down_proj.scales": "model.safetensors",
221
+ "model.layers.16.mlp.down_proj.weight": "model.safetensors",
222
+ "model.layers.16.mlp.gate_proj.biases": "model.safetensors",
223
+ "model.layers.16.mlp.gate_proj.scales": "model.safetensors",
224
+ "model.layers.16.mlp.gate_proj.weight": "model.safetensors",
225
+ "model.layers.16.mlp.up_proj.biases": "model.safetensors",
226
+ "model.layers.16.mlp.up_proj.scales": "model.safetensors",
227
+ "model.layers.16.mlp.up_proj.weight": "model.safetensors",
228
+ "model.layers.16.post_attention_layernorm.weight": "model.safetensors",
229
+ "model.layers.16.self_attn.k_proj.bias": "model.safetensors",
230
+ "model.layers.16.self_attn.k_proj.biases": "model.safetensors",
231
+ "model.layers.16.self_attn.k_proj.scales": "model.safetensors",
232
+ "model.layers.16.self_attn.k_proj.weight": "model.safetensors",
233
+ "model.layers.16.self_attn.o_proj.biases": "model.safetensors",
234
+ "model.layers.16.self_attn.o_proj.scales": "model.safetensors",
235
+ "model.layers.16.self_attn.o_proj.weight": "model.safetensors",
236
+ "model.layers.16.self_attn.q_proj.bias": "model.safetensors",
237
+ "model.layers.16.self_attn.q_proj.biases": "model.safetensors",
238
+ "model.layers.16.self_attn.q_proj.scales": "model.safetensors",
239
+ "model.layers.16.self_attn.q_proj.weight": "model.safetensors",
240
+ "model.layers.16.self_attn.v_proj.bias": "model.safetensors",
241
+ "model.layers.16.self_attn.v_proj.biases": "model.safetensors",
242
+ "model.layers.16.self_attn.v_proj.scales": "model.safetensors",
243
+ "model.layers.16.self_attn.v_proj.weight": "model.safetensors",
244
+ "model.layers.17.input_layernorm.weight": "model.safetensors",
245
+ "model.layers.17.mlp.down_proj.biases": "model.safetensors",
246
+ "model.layers.17.mlp.down_proj.scales": "model.safetensors",
247
+ "model.layers.17.mlp.down_proj.weight": "model.safetensors",
248
+ "model.layers.17.mlp.gate_proj.biases": "model.safetensors",
249
+ "model.layers.17.mlp.gate_proj.scales": "model.safetensors",
250
+ "model.layers.17.mlp.gate_proj.weight": "model.safetensors",
251
+ "model.layers.17.mlp.up_proj.biases": "model.safetensors",
252
+ "model.layers.17.mlp.up_proj.scales": "model.safetensors",
253
+ "model.layers.17.mlp.up_proj.weight": "model.safetensors",
254
+ "model.layers.17.post_attention_layernorm.weight": "model.safetensors",
255
+ "model.layers.17.self_attn.k_proj.bias": "model.safetensors",
256
+ "model.layers.17.self_attn.k_proj.biases": "model.safetensors",
257
+ "model.layers.17.self_attn.k_proj.scales": "model.safetensors",
258
+ "model.layers.17.self_attn.k_proj.weight": "model.safetensors",
259
+ "model.layers.17.self_attn.o_proj.biases": "model.safetensors",
260
+ "model.layers.17.self_attn.o_proj.scales": "model.safetensors",
261
+ "model.layers.17.self_attn.o_proj.weight": "model.safetensors",
262
+ "model.layers.17.self_attn.q_proj.bias": "model.safetensors",
263
+ "model.layers.17.self_attn.q_proj.biases": "model.safetensors",
264
+ "model.layers.17.self_attn.q_proj.scales": "model.safetensors",
265
+ "model.layers.17.self_attn.q_proj.weight": "model.safetensors",
266
+ "model.layers.17.self_attn.v_proj.bias": "model.safetensors",
267
+ "model.layers.17.self_attn.v_proj.biases": "model.safetensors",
268
+ "model.layers.17.self_attn.v_proj.scales": "model.safetensors",
269
+ "model.layers.17.self_attn.v_proj.weight": "model.safetensors",
270
+ "model.layers.18.input_layernorm.weight": "model.safetensors",
271
+ "model.layers.18.mlp.down_proj.biases": "model.safetensors",
272
+ "model.layers.18.mlp.down_proj.scales": "model.safetensors",
273
+ "model.layers.18.mlp.down_proj.weight": "model.safetensors",
274
+ "model.layers.18.mlp.gate_proj.biases": "model.safetensors",
275
+ "model.layers.18.mlp.gate_proj.scales": "model.safetensors",
276
+ "model.layers.18.mlp.gate_proj.weight": "model.safetensors",
277
+ "model.layers.18.mlp.up_proj.biases": "model.safetensors",
278
+ "model.layers.18.mlp.up_proj.scales": "model.safetensors",
279
+ "model.layers.18.mlp.up_proj.weight": "model.safetensors",
280
+ "model.layers.18.post_attention_layernorm.weight": "model.safetensors",
281
+ "model.layers.18.self_attn.k_proj.bias": "model.safetensors",
282
+ "model.layers.18.self_attn.k_proj.biases": "model.safetensors",
283
+ "model.layers.18.self_attn.k_proj.scales": "model.safetensors",
284
+ "model.layers.18.self_attn.k_proj.weight": "model.safetensors",
285
+ "model.layers.18.self_attn.o_proj.biases": "model.safetensors",
286
+ "model.layers.18.self_attn.o_proj.scales": "model.safetensors",
287
+ "model.layers.18.self_attn.o_proj.weight": "model.safetensors",
288
+ "model.layers.18.self_attn.q_proj.bias": "model.safetensors",
289
+ "model.layers.18.self_attn.q_proj.biases": "model.safetensors",
290
+ "model.layers.18.self_attn.q_proj.scales": "model.safetensors",
291
+ "model.layers.18.self_attn.q_proj.weight": "model.safetensors",
292
+ "model.layers.18.self_attn.v_proj.bias": "model.safetensors",
293
+ "model.layers.18.self_attn.v_proj.biases": "model.safetensors",
294
+ "model.layers.18.self_attn.v_proj.scales": "model.safetensors",
295
+ "model.layers.18.self_attn.v_proj.weight": "model.safetensors",
296
+ "model.layers.19.input_layernorm.weight": "model.safetensors",
297
+ "model.layers.19.mlp.down_proj.biases": "model.safetensors",
298
+ "model.layers.19.mlp.down_proj.scales": "model.safetensors",
299
+ "model.layers.19.mlp.down_proj.weight": "model.safetensors",
300
+ "model.layers.19.mlp.gate_proj.biases": "model.safetensors",
301
+ "model.layers.19.mlp.gate_proj.scales": "model.safetensors",
302
+ "model.layers.19.mlp.gate_proj.weight": "model.safetensors",
303
+ "model.layers.19.mlp.up_proj.biases": "model.safetensors",
304
+ "model.layers.19.mlp.up_proj.scales": "model.safetensors",
305
+ "model.layers.19.mlp.up_proj.weight": "model.safetensors",
306
+ "model.layers.19.post_attention_layernorm.weight": "model.safetensors",
307
+ "model.layers.19.self_attn.k_proj.bias": "model.safetensors",
308
+ "model.layers.19.self_attn.k_proj.biases": "model.safetensors",
309
+ "model.layers.19.self_attn.k_proj.scales": "model.safetensors",
310
+ "model.layers.19.self_attn.k_proj.weight": "model.safetensors",
311
+ "model.layers.19.self_attn.o_proj.biases": "model.safetensors",
312
+ "model.layers.19.self_attn.o_proj.scales": "model.safetensors",
313
+ "model.layers.19.self_attn.o_proj.weight": "model.safetensors",
314
+ "model.layers.19.self_attn.q_proj.bias": "model.safetensors",
315
+ "model.layers.19.self_attn.q_proj.biases": "model.safetensors",
316
+ "model.layers.19.self_attn.q_proj.scales": "model.safetensors",
317
+ "model.layers.19.self_attn.q_proj.weight": "model.safetensors",
318
+ "model.layers.19.self_attn.v_proj.bias": "model.safetensors",
319
+ "model.layers.19.self_attn.v_proj.biases": "model.safetensors",
320
+ "model.layers.19.self_attn.v_proj.scales": "model.safetensors",
321
+ "model.layers.19.self_attn.v_proj.weight": "model.safetensors",
322
+ "model.layers.2.input_layernorm.weight": "model.safetensors",
323
+ "model.layers.2.mlp.down_proj.biases": "model.safetensors",
324
+ "model.layers.2.mlp.down_proj.scales": "model.safetensors",
325
+ "model.layers.2.mlp.down_proj.weight": "model.safetensors",
326
+ "model.layers.2.mlp.gate_proj.biases": "model.safetensors",
327
+ "model.layers.2.mlp.gate_proj.scales": "model.safetensors",
328
+ "model.layers.2.mlp.gate_proj.weight": "model.safetensors",
329
+ "model.layers.2.mlp.up_proj.biases": "model.safetensors",
330
+ "model.layers.2.mlp.up_proj.scales": "model.safetensors",
331
+ "model.layers.2.mlp.up_proj.weight": "model.safetensors",
332
+ "model.layers.2.post_attention_layernorm.weight": "model.safetensors",
333
+ "model.layers.2.self_attn.k_proj.bias": "model.safetensors",
334
+ "model.layers.2.self_attn.k_proj.biases": "model.safetensors",
335
+ "model.layers.2.self_attn.k_proj.scales": "model.safetensors",
336
+ "model.layers.2.self_attn.k_proj.weight": "model.safetensors",
337
+ "model.layers.2.self_attn.o_proj.biases": "model.safetensors",
338
+ "model.layers.2.self_attn.o_proj.scales": "model.safetensors",
339
+ "model.layers.2.self_attn.o_proj.weight": "model.safetensors",
340
+ "model.layers.2.self_attn.q_proj.bias": "model.safetensors",
341
+ "model.layers.2.self_attn.q_proj.biases": "model.safetensors",
342
+ "model.layers.2.self_attn.q_proj.scales": "model.safetensors",
343
+ "model.layers.2.self_attn.q_proj.weight": "model.safetensors",
344
+ "model.layers.2.self_attn.v_proj.bias": "model.safetensors",
345
+ "model.layers.2.self_attn.v_proj.biases": "model.safetensors",
346
+ "model.layers.2.self_attn.v_proj.scales": "model.safetensors",
347
+ "model.layers.2.self_attn.v_proj.weight": "model.safetensors",
348
+ "model.layers.20.input_layernorm.weight": "model.safetensors",
349
+ "model.layers.20.mlp.down_proj.biases": "model.safetensors",
350
+ "model.layers.20.mlp.down_proj.scales": "model.safetensors",
351
+ "model.layers.20.mlp.down_proj.weight": "model.safetensors",
352
+ "model.layers.20.mlp.gate_proj.biases": "model.safetensors",
353
+ "model.layers.20.mlp.gate_proj.scales": "model.safetensors",
354
+ "model.layers.20.mlp.gate_proj.weight": "model.safetensors",
355
+ "model.layers.20.mlp.up_proj.biases": "model.safetensors",
356
+ "model.layers.20.mlp.up_proj.scales": "model.safetensors",
357
+ "model.layers.20.mlp.up_proj.weight": "model.safetensors",
358
+ "model.layers.20.post_attention_layernorm.weight": "model.safetensors",
359
+ "model.layers.20.self_attn.k_proj.bias": "model.safetensors",
360
+ "model.layers.20.self_attn.k_proj.biases": "model.safetensors",
361
+ "model.layers.20.self_attn.k_proj.scales": "model.safetensors",
362
+ "model.layers.20.self_attn.k_proj.weight": "model.safetensors",
363
+ "model.layers.20.self_attn.o_proj.biases": "model.safetensors",
364
+ "model.layers.20.self_attn.o_proj.scales": "model.safetensors",
365
+ "model.layers.20.self_attn.o_proj.weight": "model.safetensors",
366
+ "model.layers.20.self_attn.q_proj.bias": "model.safetensors",
367
+ "model.layers.20.self_attn.q_proj.biases": "model.safetensors",
368
+ "model.layers.20.self_attn.q_proj.scales": "model.safetensors",
369
+ "model.layers.20.self_attn.q_proj.weight": "model.safetensors",
370
+ "model.layers.20.self_attn.v_proj.bias": "model.safetensors",
371
+ "model.layers.20.self_attn.v_proj.biases": "model.safetensors",
372
+ "model.layers.20.self_attn.v_proj.scales": "model.safetensors",
373
+ "model.layers.20.self_attn.v_proj.weight": "model.safetensors",
374
+ "model.layers.21.input_layernorm.weight": "model.safetensors",
375
+ "model.layers.21.mlp.down_proj.biases": "model.safetensors",
376
+ "model.layers.21.mlp.down_proj.scales": "model.safetensors",
377
+ "model.layers.21.mlp.down_proj.weight": "model.safetensors",
378
+ "model.layers.21.mlp.gate_proj.biases": "model.safetensors",
379
+ "model.layers.21.mlp.gate_proj.scales": "model.safetensors",
380
+ "model.layers.21.mlp.gate_proj.weight": "model.safetensors",
381
+ "model.layers.21.mlp.up_proj.biases": "model.safetensors",
382
+ "model.layers.21.mlp.up_proj.scales": "model.safetensors",
383
+ "model.layers.21.mlp.up_proj.weight": "model.safetensors",
384
+ "model.layers.21.post_attention_layernorm.weight": "model.safetensors",
385
+ "model.layers.21.self_attn.k_proj.bias": "model.safetensors",
386
+ "model.layers.21.self_attn.k_proj.biases": "model.safetensors",
387
+ "model.layers.21.self_attn.k_proj.scales": "model.safetensors",
388
+ "model.layers.21.self_attn.k_proj.weight": "model.safetensors",
389
+ "model.layers.21.self_attn.o_proj.biases": "model.safetensors",
390
+ "model.layers.21.self_attn.o_proj.scales": "model.safetensors",
391
+ "model.layers.21.self_attn.o_proj.weight": "model.safetensors",
392
+ "model.layers.21.self_attn.q_proj.bias": "model.safetensors",
393
+ "model.layers.21.self_attn.q_proj.biases": "model.safetensors",
394
+ "model.layers.21.self_attn.q_proj.scales": "model.safetensors",
395
+ "model.layers.21.self_attn.q_proj.weight": "model.safetensors",
396
+ "model.layers.21.self_attn.v_proj.bias": "model.safetensors",
397
+ "model.layers.21.self_attn.v_proj.biases": "model.safetensors",
398
+ "model.layers.21.self_attn.v_proj.scales": "model.safetensors",
399
+ "model.layers.21.self_attn.v_proj.weight": "model.safetensors",
400
+ "model.layers.22.input_layernorm.weight": "model.safetensors",
401
+ "model.layers.22.mlp.down_proj.biases": "model.safetensors",
402
+ "model.layers.22.mlp.down_proj.scales": "model.safetensors",
403
+ "model.layers.22.mlp.down_proj.weight": "model.safetensors",
404
+ "model.layers.22.mlp.gate_proj.biases": "model.safetensors",
405
+ "model.layers.22.mlp.gate_proj.scales": "model.safetensors",
406
+ "model.layers.22.mlp.gate_proj.weight": "model.safetensors",
407
+ "model.layers.22.mlp.up_proj.biases": "model.safetensors",
408
+ "model.layers.22.mlp.up_proj.scales": "model.safetensors",
409
+ "model.layers.22.mlp.up_proj.weight": "model.safetensors",
410
+ "model.layers.22.post_attention_layernorm.weight": "model.safetensors",
411
+ "model.layers.22.self_attn.k_proj.bias": "model.safetensors",
412
+ "model.layers.22.self_attn.k_proj.biases": "model.safetensors",
413
+ "model.layers.22.self_attn.k_proj.scales": "model.safetensors",
414
+ "model.layers.22.self_attn.k_proj.weight": "model.safetensors",
415
+ "model.layers.22.self_attn.o_proj.biases": "model.safetensors",
416
+ "model.layers.22.self_attn.o_proj.scales": "model.safetensors",
417
+ "model.layers.22.self_attn.o_proj.weight": "model.safetensors",
418
+ "model.layers.22.self_attn.q_proj.bias": "model.safetensors",
419
+ "model.layers.22.self_attn.q_proj.biases": "model.safetensors",
420
+ "model.layers.22.self_attn.q_proj.scales": "model.safetensors",
421
+ "model.layers.22.self_attn.q_proj.weight": "model.safetensors",
422
+ "model.layers.22.self_attn.v_proj.bias": "model.safetensors",
423
+ "model.layers.22.self_attn.v_proj.biases": "model.safetensors",
424
+ "model.layers.22.self_attn.v_proj.scales": "model.safetensors",
425
+ "model.layers.22.self_attn.v_proj.weight": "model.safetensors",
426
+ "model.layers.23.input_layernorm.weight": "model.safetensors",
427
+ "model.layers.23.mlp.down_proj.biases": "model.safetensors",
428
+ "model.layers.23.mlp.down_proj.scales": "model.safetensors",
429
+ "model.layers.23.mlp.down_proj.weight": "model.safetensors",
430
+ "model.layers.23.mlp.gate_proj.biases": "model.safetensors",
431
+ "model.layers.23.mlp.gate_proj.scales": "model.safetensors",
432
+ "model.layers.23.mlp.gate_proj.weight": "model.safetensors",
433
+ "model.layers.23.mlp.up_proj.biases": "model.safetensors",
434
+ "model.layers.23.mlp.up_proj.scales": "model.safetensors",
435
+ "model.layers.23.mlp.up_proj.weight": "model.safetensors",
436
+ "model.layers.23.post_attention_layernorm.weight": "model.safetensors",
437
+ "model.layers.23.self_attn.k_proj.bias": "model.safetensors",
438
+ "model.layers.23.self_attn.k_proj.biases": "model.safetensors",
439
+ "model.layers.23.self_attn.k_proj.scales": "model.safetensors",
440
+ "model.layers.23.self_attn.k_proj.weight": "model.safetensors",
441
+ "model.layers.23.self_attn.o_proj.biases": "model.safetensors",
442
+ "model.layers.23.self_attn.o_proj.scales": "model.safetensors",
443
+ "model.layers.23.self_attn.o_proj.weight": "model.safetensors",
444
+ "model.layers.23.self_attn.q_proj.bias": "model.safetensors",
445
+ "model.layers.23.self_attn.q_proj.biases": "model.safetensors",
446
+ "model.layers.23.self_attn.q_proj.scales": "model.safetensors",
447
+ "model.layers.23.self_attn.q_proj.weight": "model.safetensors",
448
+ "model.layers.23.self_attn.v_proj.bias": "model.safetensors",
449
+ "model.layers.23.self_attn.v_proj.biases": "model.safetensors",
450
+ "model.layers.23.self_attn.v_proj.scales": "model.safetensors",
451
+ "model.layers.23.self_attn.v_proj.weight": "model.safetensors",
452
+ "model.layers.24.input_layernorm.weight": "model.safetensors",
453
+ "model.layers.24.mlp.down_proj.biases": "model.safetensors",
454
+ "model.layers.24.mlp.down_proj.scales": "model.safetensors",
455
+ "model.layers.24.mlp.down_proj.weight": "model.safetensors",
456
+ "model.layers.24.mlp.gate_proj.biases": "model.safetensors",
457
+ "model.layers.24.mlp.gate_proj.scales": "model.safetensors",
458
+ "model.layers.24.mlp.gate_proj.weight": "model.safetensors",
459
+ "model.layers.24.mlp.up_proj.biases": "model.safetensors",
460
+ "model.layers.24.mlp.up_proj.scales": "model.safetensors",
461
+ "model.layers.24.mlp.up_proj.weight": "model.safetensors",
462
+ "model.layers.24.post_attention_layernorm.weight": "model.safetensors",
463
+ "model.layers.24.self_attn.k_proj.bias": "model.safetensors",
464
+ "model.layers.24.self_attn.k_proj.biases": "model.safetensors",
465
+ "model.layers.24.self_attn.k_proj.scales": "model.safetensors",
466
+ "model.layers.24.self_attn.k_proj.weight": "model.safetensors",
467
+ "model.layers.24.self_attn.o_proj.biases": "model.safetensors",
468
+ "model.layers.24.self_attn.o_proj.scales": "model.safetensors",
469
+ "model.layers.24.self_attn.o_proj.weight": "model.safetensors",
470
+ "model.layers.24.self_attn.q_proj.bias": "model.safetensors",
471
+ "model.layers.24.self_attn.q_proj.biases": "model.safetensors",
472
+ "model.layers.24.self_attn.q_proj.scales": "model.safetensors",
473
+ "model.layers.24.self_attn.q_proj.weight": "model.safetensors",
474
+ "model.layers.24.self_attn.v_proj.bias": "model.safetensors",
475
+ "model.layers.24.self_attn.v_proj.biases": "model.safetensors",
476
+ "model.layers.24.self_attn.v_proj.scales": "model.safetensors",
477
+ "model.layers.24.self_attn.v_proj.weight": "model.safetensors",
478
+ "model.layers.25.input_layernorm.weight": "model.safetensors",
479
+ "model.layers.25.mlp.down_proj.biases": "model.safetensors",
480
+ "model.layers.25.mlp.down_proj.scales": "model.safetensors",
481
+ "model.layers.25.mlp.down_proj.weight": "model.safetensors",
482
+ "model.layers.25.mlp.gate_proj.biases": "model.safetensors",
483
+ "model.layers.25.mlp.gate_proj.scales": "model.safetensors",
484
+ "model.layers.25.mlp.gate_proj.weight": "model.safetensors",
485
+ "model.layers.25.mlp.up_proj.biases": "model.safetensors",
486
+ "model.layers.25.mlp.up_proj.scales": "model.safetensors",
487
+ "model.layers.25.mlp.up_proj.weight": "model.safetensors",
488
+ "model.layers.25.post_attention_layernorm.weight": "model.safetensors",
489
+ "model.layers.25.self_attn.k_proj.bias": "model.safetensors",
490
+ "model.layers.25.self_attn.k_proj.biases": "model.safetensors",
491
+ "model.layers.25.self_attn.k_proj.scales": "model.safetensors",
492
+ "model.layers.25.self_attn.k_proj.weight": "model.safetensors",
493
+ "model.layers.25.self_attn.o_proj.biases": "model.safetensors",
494
+ "model.layers.25.self_attn.o_proj.scales": "model.safetensors",
495
+ "model.layers.25.self_attn.o_proj.weight": "model.safetensors",
496
+ "model.layers.25.self_attn.q_proj.bias": "model.safetensors",
497
+ "model.layers.25.self_attn.q_proj.biases": "model.safetensors",
498
+ "model.layers.25.self_attn.q_proj.scales": "model.safetensors",
499
+ "model.layers.25.self_attn.q_proj.weight": "model.safetensors",
500
+ "model.layers.25.self_attn.v_proj.bias": "model.safetensors",
501
+ "model.layers.25.self_attn.v_proj.biases": "model.safetensors",
502
+ "model.layers.25.self_attn.v_proj.scales": "model.safetensors",
503
+ "model.layers.25.self_attn.v_proj.weight": "model.safetensors",
504
+ "model.layers.26.input_layernorm.weight": "model.safetensors",
505
+ "model.layers.26.mlp.down_proj.biases": "model.safetensors",
506
+ "model.layers.26.mlp.down_proj.scales": "model.safetensors",
507
+ "model.layers.26.mlp.down_proj.weight": "model.safetensors",
508
+ "model.layers.26.mlp.gate_proj.biases": "model.safetensors",
509
+ "model.layers.26.mlp.gate_proj.scales": "model.safetensors",
510
+ "model.layers.26.mlp.gate_proj.weight": "model.safetensors",
511
+ "model.layers.26.mlp.up_proj.biases": "model.safetensors",
512
+ "model.layers.26.mlp.up_proj.scales": "model.safetensors",
513
+ "model.layers.26.mlp.up_proj.weight": "model.safetensors",
514
+ "model.layers.26.post_attention_layernorm.weight": "model.safetensors",
515
+ "model.layers.26.self_attn.k_proj.bias": "model.safetensors",
516
+ "model.layers.26.self_attn.k_proj.biases": "model.safetensors",
517
+ "model.layers.26.self_attn.k_proj.scales": "model.safetensors",
518
+ "model.layers.26.self_attn.k_proj.weight": "model.safetensors",
519
+ "model.layers.26.self_attn.o_proj.biases": "model.safetensors",
520
+ "model.layers.26.self_attn.o_proj.scales": "model.safetensors",
521
+ "model.layers.26.self_attn.o_proj.weight": "model.safetensors",
522
+ "model.layers.26.self_attn.q_proj.bias": "model.safetensors",
523
+ "model.layers.26.self_attn.q_proj.biases": "model.safetensors",
524
+ "model.layers.26.self_attn.q_proj.scales": "model.safetensors",
525
+ "model.layers.26.self_attn.q_proj.weight": "model.safetensors",
526
+ "model.layers.26.self_attn.v_proj.bias": "model.safetensors",
527
+ "model.layers.26.self_attn.v_proj.biases": "model.safetensors",
528
+ "model.layers.26.self_attn.v_proj.scales": "model.safetensors",
529
+ "model.layers.26.self_attn.v_proj.weight": "model.safetensors",
530
+ "model.layers.27.input_layernorm.weight": "model.safetensors",
531
+ "model.layers.27.mlp.down_proj.biases": "model.safetensors",
532
+ "model.layers.27.mlp.down_proj.scales": "model.safetensors",
533
+ "model.layers.27.mlp.down_proj.weight": "model.safetensors",
534
+ "model.layers.27.mlp.gate_proj.biases": "model.safetensors",
535
+ "model.layers.27.mlp.gate_proj.scales": "model.safetensors",
536
+ "model.layers.27.mlp.gate_proj.weight": "model.safetensors",
537
+ "model.layers.27.mlp.up_proj.biases": "model.safetensors",
538
+ "model.layers.27.mlp.up_proj.scales": "model.safetensors",
539
+ "model.layers.27.mlp.up_proj.weight": "model.safetensors",
540
+ "model.layers.27.post_attention_layernorm.weight": "model.safetensors",
541
+ "model.layers.27.self_attn.k_proj.bias": "model.safetensors",
542
+ "model.layers.27.self_attn.k_proj.biases": "model.safetensors",
543
+ "model.layers.27.self_attn.k_proj.scales": "model.safetensors",
544
+ "model.layers.27.self_attn.k_proj.weight": "model.safetensors",
545
+ "model.layers.27.self_attn.o_proj.biases": "model.safetensors",
546
+ "model.layers.27.self_attn.o_proj.scales": "model.safetensors",
547
+ "model.layers.27.self_attn.o_proj.weight": "model.safetensors",
548
+ "model.layers.27.self_attn.q_proj.bias": "model.safetensors",
549
+ "model.layers.27.self_attn.q_proj.biases": "model.safetensors",
550
+ "model.layers.27.self_attn.q_proj.scales": "model.safetensors",
551
+ "model.layers.27.self_attn.q_proj.weight": "model.safetensors",
552
+ "model.layers.27.self_attn.v_proj.bias": "model.safetensors",
553
+ "model.layers.27.self_attn.v_proj.biases": "model.safetensors",
554
+ "model.layers.27.self_attn.v_proj.scales": "model.safetensors",
555
+ "model.layers.27.self_attn.v_proj.weight": "model.safetensors",
556
+ "model.layers.3.input_layernorm.weight": "model.safetensors",
557
+ "model.layers.3.mlp.down_proj.biases": "model.safetensors",
558
+ "model.layers.3.mlp.down_proj.scales": "model.safetensors",
559
+ "model.layers.3.mlp.down_proj.weight": "model.safetensors",
560
+ "model.layers.3.mlp.gate_proj.biases": "model.safetensors",
561
+ "model.layers.3.mlp.gate_proj.scales": "model.safetensors",
562
+ "model.layers.3.mlp.gate_proj.weight": "model.safetensors",
563
+ "model.layers.3.mlp.up_proj.biases": "model.safetensors",
564
+ "model.layers.3.mlp.up_proj.scales": "model.safetensors",
565
+ "model.layers.3.mlp.up_proj.weight": "model.safetensors",
566
+ "model.layers.3.post_attention_layernorm.weight": "model.safetensors",
567
+ "model.layers.3.self_attn.k_proj.bias": "model.safetensors",
568
+ "model.layers.3.self_attn.k_proj.biases": "model.safetensors",
569
+ "model.layers.3.self_attn.k_proj.scales": "model.safetensors",
570
+ "model.layers.3.self_attn.k_proj.weight": "model.safetensors",
571
+ "model.layers.3.self_attn.o_proj.biases": "model.safetensors",
572
+ "model.layers.3.self_attn.o_proj.scales": "model.safetensors",
573
+ "model.layers.3.self_attn.o_proj.weight": "model.safetensors",
574
+ "model.layers.3.self_attn.q_proj.bias": "model.safetensors",
575
+ "model.layers.3.self_attn.q_proj.biases": "model.safetensors",
576
+ "model.layers.3.self_attn.q_proj.scales": "model.safetensors",
577
+ "model.layers.3.self_attn.q_proj.weight": "model.safetensors",
578
+ "model.layers.3.self_attn.v_proj.bias": "model.safetensors",
579
+ "model.layers.3.self_attn.v_proj.biases": "model.safetensors",
580
+ "model.layers.3.self_attn.v_proj.scales": "model.safetensors",
581
+ "model.layers.3.self_attn.v_proj.weight": "model.safetensors",
582
+ "model.layers.4.input_layernorm.weight": "model.safetensors",
583
+ "model.layers.4.mlp.down_proj.biases": "model.safetensors",
584
+ "model.layers.4.mlp.down_proj.scales": "model.safetensors",
585
+ "model.layers.4.mlp.down_proj.weight": "model.safetensors",
586
+ "model.layers.4.mlp.gate_proj.biases": "model.safetensors",
587
+ "model.layers.4.mlp.gate_proj.scales": "model.safetensors",
588
+ "model.layers.4.mlp.gate_proj.weight": "model.safetensors",
589
+ "model.layers.4.mlp.up_proj.biases": "model.safetensors",
590
+ "model.layers.4.mlp.up_proj.scales": "model.safetensors",
591
+ "model.layers.4.mlp.up_proj.weight": "model.safetensors",
592
+ "model.layers.4.post_attention_layernorm.weight": "model.safetensors",
593
+ "model.layers.4.self_attn.k_proj.bias": "model.safetensors",
594
+ "model.layers.4.self_attn.k_proj.biases": "model.safetensors",
595
+ "model.layers.4.self_attn.k_proj.scales": "model.safetensors",
596
+ "model.layers.4.self_attn.k_proj.weight": "model.safetensors",
597
+ "model.layers.4.self_attn.o_proj.biases": "model.safetensors",
598
+ "model.layers.4.self_attn.o_proj.scales": "model.safetensors",
599
+ "model.layers.4.self_attn.o_proj.weight": "model.safetensors",
600
+ "model.layers.4.self_attn.q_proj.bias": "model.safetensors",
601
+ "model.layers.4.self_attn.q_proj.biases": "model.safetensors",
602
+ "model.layers.4.self_attn.q_proj.scales": "model.safetensors",
603
+ "model.layers.4.self_attn.q_proj.weight": "model.safetensors",
604
+ "model.layers.4.self_attn.v_proj.bias": "model.safetensors",
605
+ "model.layers.4.self_attn.v_proj.biases": "model.safetensors",
606
+ "model.layers.4.self_attn.v_proj.scales": "model.safetensors",
607
+ "model.layers.4.self_attn.v_proj.weight": "model.safetensors",
608
+ "model.layers.5.input_layernorm.weight": "model.safetensors",
609
+ "model.layers.5.mlp.down_proj.biases": "model.safetensors",
610
+ "model.layers.5.mlp.down_proj.scales": "model.safetensors",
611
+ "model.layers.5.mlp.down_proj.weight": "model.safetensors",
612
+ "model.layers.5.mlp.gate_proj.biases": "model.safetensors",
613
+ "model.layers.5.mlp.gate_proj.scales": "model.safetensors",
614
+ "model.layers.5.mlp.gate_proj.weight": "model.safetensors",
615
+ "model.layers.5.mlp.up_proj.biases": "model.safetensors",
616
+ "model.layers.5.mlp.up_proj.scales": "model.safetensors",
617
+ "model.layers.5.mlp.up_proj.weight": "model.safetensors",
618
+ "model.layers.5.post_attention_layernorm.weight": "model.safetensors",
619
+ "model.layers.5.self_attn.k_proj.bias": "model.safetensors",
620
+ "model.layers.5.self_attn.k_proj.biases": "model.safetensors",
621
+ "model.layers.5.self_attn.k_proj.scales": "model.safetensors",
622
+ "model.layers.5.self_attn.k_proj.weight": "model.safetensors",
623
+ "model.layers.5.self_attn.o_proj.biases": "model.safetensors",
624
+ "model.layers.5.self_attn.o_proj.scales": "model.safetensors",
625
+ "model.layers.5.self_attn.o_proj.weight": "model.safetensors",
626
+ "model.layers.5.self_attn.q_proj.bias": "model.safetensors",
627
+ "model.layers.5.self_attn.q_proj.biases": "model.safetensors",
628
+ "model.layers.5.self_attn.q_proj.scales": "model.safetensors",
629
+ "model.layers.5.self_attn.q_proj.weight": "model.safetensors",
630
+ "model.layers.5.self_attn.v_proj.bias": "model.safetensors",
631
+ "model.layers.5.self_attn.v_proj.biases": "model.safetensors",
632
+ "model.layers.5.self_attn.v_proj.scales": "model.safetensors",
633
+ "model.layers.5.self_attn.v_proj.weight": "model.safetensors",
634
+ "model.layers.6.input_layernorm.weight": "model.safetensors",
635
+ "model.layers.6.mlp.down_proj.biases": "model.safetensors",
636
+ "model.layers.6.mlp.down_proj.scales": "model.safetensors",
637
+ "model.layers.6.mlp.down_proj.weight": "model.safetensors",
638
+ "model.layers.6.mlp.gate_proj.biases": "model.safetensors",
639
+ "model.layers.6.mlp.gate_proj.scales": "model.safetensors",
640
+ "model.layers.6.mlp.gate_proj.weight": "model.safetensors",
641
+ "model.layers.6.mlp.up_proj.biases": "model.safetensors",
642
+ "model.layers.6.mlp.up_proj.scales": "model.safetensors",
643
+ "model.layers.6.mlp.up_proj.weight": "model.safetensors",
644
+ "model.layers.6.post_attention_layernorm.weight": "model.safetensors",
645
+ "model.layers.6.self_attn.k_proj.bias": "model.safetensors",
646
+ "model.layers.6.self_attn.k_proj.biases": "model.safetensors",
647
+ "model.layers.6.self_attn.k_proj.scales": "model.safetensors",
648
+ "model.layers.6.self_attn.k_proj.weight": "model.safetensors",
649
+ "model.layers.6.self_attn.o_proj.biases": "model.safetensors",
650
+ "model.layers.6.self_attn.o_proj.scales": "model.safetensors",
651
+ "model.layers.6.self_attn.o_proj.weight": "model.safetensors",
652
+ "model.layers.6.self_attn.q_proj.bias": "model.safetensors",
653
+ "model.layers.6.self_attn.q_proj.biases": "model.safetensors",
654
+ "model.layers.6.self_attn.q_proj.scales": "model.safetensors",
655
+ "model.layers.6.self_attn.q_proj.weight": "model.safetensors",
656
+ "model.layers.6.self_attn.v_proj.bias": "model.safetensors",
657
+ "model.layers.6.self_attn.v_proj.biases": "model.safetensors",
658
+ "model.layers.6.self_attn.v_proj.scales": "model.safetensors",
659
+ "model.layers.6.self_attn.v_proj.weight": "model.safetensors",
660
+ "model.layers.7.input_layernorm.weight": "model.safetensors",
661
+ "model.layers.7.mlp.down_proj.biases": "model.safetensors",
662
+ "model.layers.7.mlp.down_proj.scales": "model.safetensors",
663
+ "model.layers.7.mlp.down_proj.weight": "model.safetensors",
664
+ "model.layers.7.mlp.gate_proj.biases": "model.safetensors",
665
+ "model.layers.7.mlp.gate_proj.scales": "model.safetensors",
666
+ "model.layers.7.mlp.gate_proj.weight": "model.safetensors",
667
+ "model.layers.7.mlp.up_proj.biases": "model.safetensors",
668
+ "model.layers.7.mlp.up_proj.scales": "model.safetensors",
669
+ "model.layers.7.mlp.up_proj.weight": "model.safetensors",
670
+ "model.layers.7.post_attention_layernorm.weight": "model.safetensors",
671
+ "model.layers.7.self_attn.k_proj.bias": "model.safetensors",
672
+ "model.layers.7.self_attn.k_proj.biases": "model.safetensors",
673
+ "model.layers.7.self_attn.k_proj.scales": "model.safetensors",
674
+ "model.layers.7.self_attn.k_proj.weight": "model.safetensors",
675
+ "model.layers.7.self_attn.o_proj.biases": "model.safetensors",
676
+ "model.layers.7.self_attn.o_proj.scales": "model.safetensors",
677
+ "model.layers.7.self_attn.o_proj.weight": "model.safetensors",
678
+ "model.layers.7.self_attn.q_proj.bias": "model.safetensors",
679
+ "model.layers.7.self_attn.q_proj.biases": "model.safetensors",
680
+ "model.layers.7.self_attn.q_proj.scales": "model.safetensors",
681
+ "model.layers.7.self_attn.q_proj.weight": "model.safetensors",
682
+ "model.layers.7.self_attn.v_proj.bias": "model.safetensors",
683
+ "model.layers.7.self_attn.v_proj.biases": "model.safetensors",
684
+ "model.layers.7.self_attn.v_proj.scales": "model.safetensors",
685
+ "model.layers.7.self_attn.v_proj.weight": "model.safetensors",
686
+ "model.layers.8.input_layernorm.weight": "model.safetensors",
687
+ "model.layers.8.mlp.down_proj.biases": "model.safetensors",
688
+ "model.layers.8.mlp.down_proj.scales": "model.safetensors",
689
+ "model.layers.8.mlp.down_proj.weight": "model.safetensors",
690
+ "model.layers.8.mlp.gate_proj.biases": "model.safetensors",
691
+ "model.layers.8.mlp.gate_proj.scales": "model.safetensors",
692
+ "model.layers.8.mlp.gate_proj.weight": "model.safetensors",
693
+ "model.layers.8.mlp.up_proj.biases": "model.safetensors",
694
+ "model.layers.8.mlp.up_proj.scales": "model.safetensors",
695
+ "model.layers.8.mlp.up_proj.weight": "model.safetensors",
696
+ "model.layers.8.post_attention_layernorm.weight": "model.safetensors",
697
+ "model.layers.8.self_attn.k_proj.bias": "model.safetensors",
698
+ "model.layers.8.self_attn.k_proj.biases": "model.safetensors",
699
+ "model.layers.8.self_attn.k_proj.scales": "model.safetensors",
700
+ "model.layers.8.self_attn.k_proj.weight": "model.safetensors",
701
+ "model.layers.8.self_attn.o_proj.biases": "model.safetensors",
702
+ "model.layers.8.self_attn.o_proj.scales": "model.safetensors",
703
+ "model.layers.8.self_attn.o_proj.weight": "model.safetensors",
704
+ "model.layers.8.self_attn.q_proj.bias": "model.safetensors",
705
+ "model.layers.8.self_attn.q_proj.biases": "model.safetensors",
706
+ "model.layers.8.self_attn.q_proj.scales": "model.safetensors",
707
+ "model.layers.8.self_attn.q_proj.weight": "model.safetensors",
708
+ "model.layers.8.self_attn.v_proj.bias": "model.safetensors",
709
+ "model.layers.8.self_attn.v_proj.biases": "model.safetensors",
710
+ "model.layers.8.self_attn.v_proj.scales": "model.safetensors",
711
+ "model.layers.8.self_attn.v_proj.weight": "model.safetensors",
712
+ "model.layers.9.input_layernorm.weight": "model.safetensors",
713
+ "model.layers.9.mlp.down_proj.biases": "model.safetensors",
714
+ "model.layers.9.mlp.down_proj.scales": "model.safetensors",
715
+ "model.layers.9.mlp.down_proj.weight": "model.safetensors",
716
+ "model.layers.9.mlp.gate_proj.biases": "model.safetensors",
717
+ "model.layers.9.mlp.gate_proj.scales": "model.safetensors",
718
+ "model.layers.9.mlp.gate_proj.weight": "model.safetensors",
719
+ "model.layers.9.mlp.up_proj.biases": "model.safetensors",
720
+ "model.layers.9.mlp.up_proj.scales": "model.safetensors",
721
+ "model.layers.9.mlp.up_proj.weight": "model.safetensors",
722
+ "model.layers.9.post_attention_layernorm.weight": "model.safetensors",
723
+ "model.layers.9.self_attn.k_proj.bias": "model.safetensors",
724
+ "model.layers.9.self_attn.k_proj.biases": "model.safetensors",
725
+ "model.layers.9.self_attn.k_proj.scales": "model.safetensors",
726
+ "model.layers.9.self_attn.k_proj.weight": "model.safetensors",
727
+ "model.layers.9.self_attn.o_proj.biases": "model.safetensors",
728
+ "model.layers.9.self_attn.o_proj.scales": "model.safetensors",
729
+ "model.layers.9.self_attn.o_proj.weight": "model.safetensors",
730
+ "model.layers.9.self_attn.q_proj.bias": "model.safetensors",
731
+ "model.layers.9.self_attn.q_proj.biases": "model.safetensors",
732
+ "model.layers.9.self_attn.q_proj.scales": "model.safetensors",
733
+ "model.layers.9.self_attn.q_proj.weight": "model.safetensors",
734
+ "model.layers.9.self_attn.v_proj.bias": "model.safetensors",
735
+ "model.layers.9.self_attn.v_proj.biases": "model.safetensors",
736
+ "model.layers.9.self_attn.v_proj.scales": "model.safetensors",
737
+ "model.layers.9.self_attn.v_proj.weight": "model.safetensors",
738
+ "model.norm.weight": "model.safetensors"
739
+ }
740
+ }
mf/backbone/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37dfca640a02fc2d757600faff32f365dbc4b367a99c4baceea98360a9561a41
3
+ size 11423259
mf/backbone/tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": null,
5
+ "clean_up_tokenization_spaces": false,
6
+ "eos_token": "<|endoftext|>",
7
+ "errors": "replace",
8
+ "is_local": true,
9
+ "local_files_only": false,
10
+ "model_max_length": 131072,
11
+ "pad_token": "<|endoftext|>",
12
+ "split_special_tokens": false,
13
+ "tokenizer_class": "Qwen2Tokenizer",
14
+ "unk_token": null
15
+ }
mf/config.json ADDED
@@ -0,0 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_type": "dots_tts",
3
+ "latent_dim": 128,
4
+ "patch_size": 4,
5
+ "cfg_droprate": 0.2,
6
+ "PatchEncoder": {
7
+ "num_layers": 24,
8
+ "num_heads": 16,
9
+ "hidden_size": 1024,
10
+ "ffn_hidden_size": 4096,
11
+ "modulation": false,
12
+ "qkv_bias": false,
13
+ "qk_norm": true,
14
+ "attn_dropout": 0.0,
15
+ "dropout": 0.1,
16
+ "norm_layer": "RMSNorm",
17
+ "alibi_bias": false,
18
+ "rotary_bias": true,
19
+ "rotary_theta": 10000.0,
20
+ "input_dim": 128,
21
+ "causal": true
22
+ },
23
+ "DiT": {
24
+ "num_layers": 18,
25
+ "num_heads": 16,
26
+ "hidden_size": 1024,
27
+ "ffn_hidden_size": 4096,
28
+ "modulation": true,
29
+ "qkv_bias": false,
30
+ "qk_norm": true,
31
+ "attn_dropout": 0.0,
32
+ "dropout": 0.0,
33
+ "norm_layer": "RMSNorm",
34
+ "alibi_bias": false,
35
+ "rotary_bias": true,
36
+ "rotary_theta": 10000.0
37
+ },
38
+ "vocoder": {
39
+ "sample_rate": 48000,
40
+ "upsample_rates": [
41
+ 10,
42
+ 6,
43
+ 4,
44
+ 2,
45
+ 2,
46
+ 2
47
+ ],
48
+ "upsample_kernel_sizes": [
49
+ 20,
50
+ 12,
51
+ 8,
52
+ 4,
53
+ 4,
54
+ 4
55
+ ],
56
+ "upsample_initial_channel": 1536,
57
+ "resblock": "1",
58
+ "resblock_kernel_sizes": [
59
+ 3,
60
+ 7,
61
+ 11
62
+ ],
63
+ "resblock_dilation_sizes": [
64
+ [
65
+ 1,
66
+ 3,
67
+ 5
68
+ ],
69
+ [
70
+ 1,
71
+ 3,
72
+ 5
73
+ ],
74
+ [
75
+ 1,
76
+ 3,
77
+ 5
78
+ ]
79
+ ],
80
+ "downsample_rates": [
81
+ 2,
82
+ 2,
83
+ 2,
84
+ 4,
85
+ 6,
86
+ 10
87
+ ],
88
+ "downsample_channels": [
89
+ 12,
90
+ 24,
91
+ 48,
92
+ 96,
93
+ 192,
94
+ 384,
95
+ 768
96
+ ],
97
+ "activation": "snakebeta",
98
+ "snake_logscale": true,
99
+ "latent_dim": 128,
100
+ "causal": true,
101
+ "mi_num_layers": 4,
102
+ "causal_encoder": true,
103
+ "use_bias_at_final": false,
104
+ "use_tanh_at_final": false
105
+ },
106
+ "fm_sigma": 0.0,
107
+ "xvec_drop_rate": 0.2,
108
+ "campplus_embedding_size": 512,
109
+ "xvec_max_audio_seconds": 10.0,
110
+ "architectures": [
111
+ "DotsTTSForConditionalGeneration"
112
+ ]
113
+ }
mf/dit/config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "num_layers": 18,
3
+ "num_heads": 16,
4
+ "hidden_size": 1024,
5
+ "ffn_hidden_size": 4096,
6
+ "head_dim": 64,
7
+ "in_dim": 1024,
8
+ "out_dim": 128,
9
+ "patch_size": 4,
10
+ "qk_norm": true,
11
+ "norm_layer": "RMSNorm",
12
+ "rms_eps": 1.1920929e-07,
13
+ "ln_eps": 1e-05,
14
+ "modulation": true,
15
+ "rotary_theta": 10000.0,
16
+ "mode": "meanflow",
17
+ "quantization": {
18
+ "group_size": 64,
19
+ "bits": 8,
20
+ "mode": "affine"
21
+ }
22
+ }
mf/dit/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c07e3a4972715e9d13fb26fc6375776a5700219ca742c539a98f33cb81b9b151
3
+ size 389815973
mf/heads/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d0776d2a8534b2788e51d675081768c41d90f3a3ad631f59b3d9be0a0aa6e04
3
+ size 18918592
mf/latent_stats.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"mean": [-0.9025294184684753, -0.6182500123977661, -0.026301059871912003, -0.06846871972084045, -0.1782952845096588, -0.5570240616798401, -0.6109894514083862, -0.38791292905807495, -0.3249254524707794, -0.2662239670753479, 0.3478159010410309, 0.03720537945628166, -0.5750355124473572, 0.03443601354956627, -0.5377659201622009, 0.22715406119823456, -0.2407034933567047, 0.28574761748313904, -0.4125779867172241, 0.059183623641729355, -0.34213384985923767, -0.5783153772354126, 0.026603808626532555, -0.0333869606256485, -0.08291488885879517, -0.3519511818885803, 0.11998001486063004, -0.43912461400032043, 0.33889240026474, -0.4224475026130676, -0.35147470235824585, -0.055791400372982025, 0.11066599935293198, -0.00278641888871789, 0.6532911658287048, -0.009556996636092663, 0.39984285831451416, 0.09787625819444656, 0.07102298736572266, 0.15442001819610596, -0.18712204694747925, 0.1712084263563156, 0.27672144770622253, -0.3803085386753082, -0.2228352427482605, -0.324790894985199, 0.08883547782897949, 0.038199909031391144, -0.27509579062461853, -0.10280673205852509, -0.30578866600990295, 0.23430660367012024, 0.024414263665676117, -0.25659969449043274, -0.02299385704100132, 0.7069130539894104, -0.405097633600235, 0.5665390491485596, -0.19767075777053833, -0.4927915036678314, 0.19865882396697998, -0.37438997626304626, 0.266941636800766, -0.5637523531913757, -0.03907559812068939, 0.15638978779315948, -0.10351831465959549, -0.44826221466064453, 0.6975453495979309, 0.27608436346054077, -0.20201721787452698, 0.26400619745254517, 0.0681535080075264, 0.11755296587944031, 0.7705864906311035, -1.3600373268127441, 0.07215584814548492, -0.13750098645687103, 0.03367617726325989, 0.17637766897678375, -0.29273325204849243, 0.3614978492259979, -0.13353696465492249, 0.3452143967151642, 0.05898723751306534, -0.27734026312828064, 0.06529057770967484, 0.36917465925216675, -0.3961399793624878, -0.03245498985052109, 0.10380517691373825, -0.24310462176799774, -0.04018634185194969, 0.3809950649738312, -0.02391395904123783, -0.34257376194000244, -0.09236539155244827, -0.2610892355442047, 0.11655107885599136, 0.40474870800971985, 0.1758791208267212, -0.6633780598640442, -0.008149964734911919, 2.0525853633880615, 0.04391917586326599, -0.13638928532600403, 0.18733453750610352, -0.026254724711179733, -0.7545398473739624, 0.5526015162467957, -0.07521389424800873, -0.3716686964035034, 0.16373921930789948, 0.7791165709495544, 0.7361481785774231, -0.3620278537273407, -0.3784199059009552, 0.46543794870376587, -0.2694196403026581, -0.2114928662776947, 0.3204426169395447, 0.29945752024650574, -0.7347325086593628, 0.47010552883148193, 0.11410360038280487, 0.6925314664840698, 0.713970959186554, -0.00318007729947567], "var": [9.1906156539917, 7.870988845825195, 3.150926113128662, 3.050175905227661, 4.262413501739502, 3.9477379322052, 5.15797758102417, 4.121795654296875, 3.4909651279449463, 5.343992233276367, 4.0427470207214355, 3.8296546936035156, 8.234746932983398, 5.330444812774658, 4.205230712890625, 4.558896541595459, 7.257613658905029, 3.0259087085723877, 3.997598886489868, 3.436048984527588, 4.351229190826416, 4.6664862632751465, 3.643401861190796, 6.883331775665283, 3.769258975982666, 12.98837947845459, 3.0210487842559814, 3.373710870742798, 7.442956924438477, 4.355553150177002, 6.233387470245361, 6.0005669593811035, 2.6840243339538574, 2.7943620681762695, 21.47194480895996, 5.7963032722473145, 15.689336776733398, 3.2821261882781982, 3.245901107788086, 3.24704647064209, 4.0499396324157715, 3.074151039123535, 3.6589696407318115, 5.024191856384277, 4.208132266998291, 3.8009533882141113, 3.4110586643218994, 4.683138370513916, 3.8099751472473145, 3.914489984512329, 3.073166847229004, 2.562713146209717, 3.7912731170654297, 7.0935869216918945, 5.117737770080566, 6.004349231719971, 3.7375590801239014, 7.916847229003906, 4.276163578033447, 4.535323619842529, 3.311223268508911, 6.089536666870117, 5.089720249176025, 5.119976997375488, 3.738840341567993, 2.8602447509765625, 4.491348743438721, 2.9860498905181885, 9.37391185760498, 2.5369112491607666, 3.158625841140747, 3.600911855697632, 2.5978822708129883, 2.470167398452759, 5.891170024871826, 9.823240280151367, 5.728444576263428, 3.254216194152832, 3.335116386413574, 3.146914005279541, 12.309706687927246, 3.7441670894622803, 5.149557113647461, 3.2937445640563965, 2.5573980808258057, 3.5558786392211914, 6.514009475708008, 5.156071662902832, 3.4441094398498535, 2.8811047077178955, 3.7099127769470215, 3.2816600799560547, 2.328526496887207, 3.4400925636291504, 2.0153448581695557, 2.9897236824035645, 2.6417579650878906, 4.0955657958984375, 6.443847179412842, 4.788559436798096, 2.873269557952881, 6.570309162139893, 4.11572265625, 169.78680419921875, 4.587249279022217, 2.8753774166107178, 22.390058517456055, 5.543135643005371, 7.11015510559082, 8.378399848937988, 2.891625165939331, 39.79545211791992, 5.232828617095947, 7.614190101623535, 5.631987571716309, 3.167414426803589, 3.900977849960327, 6.806697845458984, 2.6496951580047607, 6.4958038330078125, 4.302042484283447, 6.472627639770508, 4.391860008239746, 2.5656657218933105, 2.0728390216827393, 10.271937370300293, 3.254828929901123, 5.5702972412109375], "shape": [128]}
mf/latent_stats.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:313b13af56d659ecf869d5f854508fcf823c8f957aefc6bc05244991abd6ffe1
3
+ size 3197
mf/llm_config.json ADDED
@@ -0,0 +1,114 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "vocab_size": 151672,
3
+ "max_position_embeddings": 131072,
4
+ "hidden_size": 1536,
5
+ "intermediate_size": 8960,
6
+ "num_hidden_layers": 28,
7
+ "num_attention_heads": 12,
8
+ "use_sliding_window": false,
9
+ "sliding_window": null,
10
+ "max_window_layers": 28,
11
+ "num_key_value_heads": 2,
12
+ "hidden_act": "silu",
13
+ "initializer_range": 0.02,
14
+ "rms_norm_eps": 1e-06,
15
+ "use_cache": true,
16
+ "rope_theta": 1000000.0,
17
+ "rope_scaling": null,
18
+ "attention_dropout": 0.0,
19
+ "layer_types": [
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
+ ],
49
+ "return_dict": true,
50
+ "output_hidden_states": false,
51
+ "torchscript": false,
52
+ "dtype": "bfloat16",
53
+ "pruned_heads": {},
54
+ "tie_word_embeddings": true,
55
+ "chunk_size_feed_forward": 0,
56
+ "is_encoder_decoder": false,
57
+ "is_decoder": false,
58
+ "cross_attention_hidden_size": null,
59
+ "add_cross_attention": false,
60
+ "tie_encoder_decoder": false,
61
+ "architectures": [
62
+ "Qwen2ForCausalLM"
63
+ ],
64
+ "finetuning_task": null,
65
+ "id2label": {
66
+ "0": "LABEL_0",
67
+ "1": "LABEL_1"
68
+ },
69
+ "label2id": {
70
+ "LABEL_0": 0,
71
+ "LABEL_1": 1
72
+ },
73
+ "task_specific_params": null,
74
+ "problem_type": null,
75
+ "tokenizer_class": null,
76
+ "prefix": null,
77
+ "bos_token_id": 151643,
78
+ "pad_token_id": null,
79
+ "eos_token_id": 151643,
80
+ "sep_token_id": null,
81
+ "decoder_start_token_id": null,
82
+ "max_length": 20,
83
+ "min_length": 0,
84
+ "do_sample": false,
85
+ "early_stopping": false,
86
+ "num_beams": 1,
87
+ "temperature": 1.0,
88
+ "top_k": 50,
89
+ "top_p": 1.0,
90
+ "typical_p": 1.0,
91
+ "repetition_penalty": 1.0,
92
+ "length_penalty": 1.0,
93
+ "no_repeat_ngram_size": 0,
94
+ "encoder_no_repeat_ngram_size": 0,
95
+ "bad_words_ids": null,
96
+ "num_return_sequences": 1,
97
+ "output_scores": false,
98
+ "return_dict_in_generate": false,
99
+ "forced_bos_token_id": null,
100
+ "forced_eos_token_id": null,
101
+ "remove_invalid_values": false,
102
+ "exponential_decay_length_penalty": null,
103
+ "suppress_tokens": null,
104
+ "begin_suppress_tokens": null,
105
+ "num_beam_groups": 1,
106
+ "diversity_penalty": 0.0,
107
+ "_name_or_path": "",
108
+ "transformers_version": "4.57.0",
109
+ "model_type": "qwen2",
110
+ "use_mrope": false,
111
+ "tf_legacy_loss": false,
112
+ "use_bfloat16": false,
113
+ "output_attentions": false
114
+ }
mf/patch_encoder/config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "quantization": {
3
+ "group_size": 64,
4
+ "bits": 8,
5
+ "mode": "affine"
6
+ }
7
+ }
mf/patch_encoder/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:412579d2a54278c1b5983c32f30dabc9f88a0586c7a2ce59360188762d987655
3
+ size 344413054
mf/resampler_48k_16k.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ea8dde91e3405e461de80b599e81bcadf1f77be682a61b28d0dff7a362c5a47
3
+ size 500
mf/speaker/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:29a51b83ccd54417e1afbb43c16539da6308709f107d24d23d36d7d63da6968c
3
+ size 29129680
mf/vocoder/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8529cf050f1a466767af38cacb47099115c19cc324a1b4a4d136ead2a21d2f83
3
+ size 273096426