Upload config.json with huggingface_hub
Browse files- config.json +59 -55
config.json
CHANGED
@@ -1,58 +1,7 @@
|
|
1 |
{
|
2 |
-
"
|
3 |
-
"lm_gen_config": {
|
4 |
-
"temp": 0.8,
|
5 |
-
"temp_text": 0.8,
|
6 |
-
"top_k": 250,
|
7 |
-
"top_k_text": 50
|
8 |
-
},
|
9 |
-
"dim": 2560,
|
10 |
-
"text_card": 48000,
|
11 |
-
"existing_text_padding_id": 3,
|
12 |
"n_q": 32,
|
13 |
"dep_q": 16,
|
14 |
-
"card": 2048,
|
15 |
-
"num_heads": 20,
|
16 |
-
"num_layers": 24,
|
17 |
-
"hidden_scale": 4.125,
|
18 |
-
"causal": true,
|
19 |
-
"layer_scale": null,
|
20 |
-
"context": 500,
|
21 |
-
"max_period": 100000,
|
22 |
-
"gating": "silu",
|
23 |
-
"norm": "rms_norm_f32",
|
24 |
-
"positional_embedding": "rope",
|
25 |
-
"depformer_dim": 1024,
|
26 |
-
"depformer_dim_feedforward": 3072,
|
27 |
-
"depformer_num_heads": 16,
|
28 |
-
"depformer_num_layers": 4,
|
29 |
-
"depformer_causal": true,
|
30 |
-
"depformer_layer_scale": null,
|
31 |
-
"depformer_multi_linear": true,
|
32 |
-
"depformer_context": 16,
|
33 |
-
"depformer_max_period": 10000,
|
34 |
-
"depformer_gating": "silu",
|
35 |
-
"depformer_pos_emb": "none",
|
36 |
-
"depformer_weights_per_step": true,
|
37 |
-
"depformer_weights_per_step_schedule": [
|
38 |
-
0,
|
39 |
-
1,
|
40 |
-
2,
|
41 |
-
3,
|
42 |
-
4,
|
43 |
-
5,
|
44 |
-
6,
|
45 |
-
7,
|
46 |
-
8,
|
47 |
-
8,
|
48 |
-
8,
|
49 |
-
8,
|
50 |
-
8,
|
51 |
-
8,
|
52 |
-
8,
|
53 |
-
8
|
54 |
-
],
|
55 |
-
"depformer_low_rank_embeddings": 128,
|
56 |
"delays": [
|
57 |
0,
|
58 |
0,
|
@@ -88,6 +37,27 @@
|
|
88 |
2,
|
89 |
2
|
90 |
],
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
91 |
"conditioners": {
|
92 |
"description": {
|
93 |
"type": "lut",
|
@@ -106,11 +76,45 @@
|
|
106 |
}
|
107 |
},
|
108 |
"fuser": {
|
|
|
|
|
109 |
"sum": [
|
110 |
"description"
|
111 |
-
]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
112 |
},
|
113 |
"mimi_name": "mimi-pytorch-e351c8d8@125.safetensors",
|
114 |
-
"
|
115 |
-
"
|
116 |
}
|
|
|
1 |
{
|
2 |
+
"card": 2048,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
"n_q": 32,
|
4 |
"dep_q": 16,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
"delays": [
|
6 |
0,
|
7 |
0,
|
|
|
37 |
2,
|
38 |
2
|
39 |
],
|
40 |
+
"dim": 2560,
|
41 |
+
"text_card": 48000,
|
42 |
+
"existing_text_padding_id": 3,
|
43 |
+
"num_heads": 20,
|
44 |
+
"num_layers": 24,
|
45 |
+
"hidden_scale": 4.125,
|
46 |
+
"causal": true,
|
47 |
+
"layer_scale": null,
|
48 |
+
"context": 1500,
|
49 |
+
"max_period": 100000.0,
|
50 |
+
"gating": "silu",
|
51 |
+
"norm": "rms_norm_f32",
|
52 |
+
"positional_embedding": "rope",
|
53 |
+
"depformer_dim": 1024,
|
54 |
+
"depformer_num_heads": 16,
|
55 |
+
"depformer_num_layers": 4,
|
56 |
+
"depformer_dim_feedforward": 3072,
|
57 |
+
"depformer_multi_linear": true,
|
58 |
+
"depformer_pos_emb": "none",
|
59 |
+
"depformer_weights_per_step": true,
|
60 |
+
"depformer_low_rank_embeddings": 128,
|
61 |
"conditioners": {
|
62 |
"description": {
|
63 |
"type": "lut",
|
|
|
76 |
}
|
77 |
},
|
78 |
"fuser": {
|
79 |
+
"cross_attention_pos_emb": false,
|
80 |
+
"cross_attention_pos_emb_scale": 1,
|
81 |
"sum": [
|
82 |
"description"
|
83 |
+
],
|
84 |
+
"prepend": [],
|
85 |
+
"cross": []
|
86 |
+
},
|
87 |
+
"cross_attention": [],
|
88 |
+
"model_id": {
|
89 |
+
"sig": "ccef4858",
|
90 |
+
"epoch": 200
|
91 |
+
},
|
92 |
+
"depformer_weights_per_step_schedule": [
|
93 |
+
0,
|
94 |
+
1,
|
95 |
+
2,
|
96 |
+
3,
|
97 |
+
4,
|
98 |
+
5,
|
99 |
+
6,
|
100 |
+
7,
|
101 |
+
8,
|
102 |
+
8,
|
103 |
+
8,
|
104 |
+
8,
|
105 |
+
8,
|
106 |
+
8,
|
107 |
+
8,
|
108 |
+
8
|
109 |
+
],
|
110 |
+
"model_type": "hibiki",
|
111 |
+
"lm_gen_config": {
|
112 |
+
"temp": 0.8,
|
113 |
+
"temp_text": 0.8,
|
114 |
+
"top_k": 250,
|
115 |
+
"top_k_text": 50
|
116 |
},
|
117 |
"mimi_name": "mimi-pytorch-e351c8d8@125.safetensors",
|
118 |
+
"tokenizer_name": "tokenizer_spm_48k_multi6_2.model",
|
119 |
+
"moshi_name": "hibiki-pytorch-ccef4858@200.safetensors"
|
120 |
}
|