Upload config.json with huggingface_hub
Browse files- config.json +52 -0
config.json
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architecture": "voxcpm",
|
| 3 |
+
"lm_config": {
|
| 4 |
+
"bos_token_id": 1,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"hidden_size": 1024,
|
| 7 |
+
"intermediate_size": 4096,
|
| 8 |
+
"max_position_embeddings": 32768,
|
| 9 |
+
"num_attention_heads": 16,
|
| 10 |
+
"num_hidden_layers": 24,
|
| 11 |
+
"num_key_value_heads": 2,
|
| 12 |
+
"rms_norm_eps": 1e-05,
|
| 13 |
+
"rope_theta": 10000,
|
| 14 |
+
"rope_scaling": {
|
| 15 |
+
"type": "longrope",
|
| 16 |
+
"long_factor": [1.0004360675811768, 1.0668443441390991, 1.1631425619125366, 1.3025742769241333, 1.5040205717086792, 1.7941505908966064, 2.2101221084594727, 2.802666664123535, 3.6389970779418945, 4.804192543029785, 6.39855432510376, 8.527148246765137, 11.277542114257812, 14.684998512268066, 18.69317054748535, 23.13019371032715, 27.72362518310547, 32.1606559753418, 36.168827056884766, 39.57627868652344, 42.32667541503906, 44.45526885986328, 46.04962921142578, 47.21482849121094, 48.05115509033203, 48.64370346069336, 49.05967712402344, 49.34980392456055, 49.551246643066406, 49.69068145751953, 49.78697967529297, 49.85338592529297],
|
| 17 |
+
"short_factor": [1.0004360675811768, 1.0668443441390991, 1.1631425619125366, 1.3025742769241333, 1.5040205717086792, 1.7941505908966064, 2.2101221084594727, 2.802666664123535, 3.6389970779418945, 4.804192543029785, 6.39855432510376, 8.527148246765137, 11.277542114257812, 14.684998512268066, 18.69317054748535, 23.13019371032715, 27.72362518310547, 32.1606559753418, 36.168827056884766, 39.57627868652344, 42.32667541503906, 44.45526885986328, 46.04962921142578, 47.21482849121094, 48.05115509033203, 48.64370346069336, 49.05967712402344, 49.34980392456055, 49.551246643066406, 49.69068145751953, 49.78697967529297, 49.85338592529297],
|
| 18 |
+
"original_max_position_embeddings": 32768
|
| 19 |
+
},
|
| 20 |
+
"vocab_size": 73448,
|
| 21 |
+
"scale_emb": 12,
|
| 22 |
+
"dim_model_base": 256,
|
| 23 |
+
"scale_depth": 1.4,
|
| 24 |
+
"use_mup": false
|
| 25 |
+
},
|
| 26 |
+
"patch_size": 2,
|
| 27 |
+
"feat_dim": 64,
|
| 28 |
+
"scalar_quantization_latent_dim": 256,
|
| 29 |
+
"scalar_quantization_scale": 9,
|
| 30 |
+
"residual_lm_num_layers": 6,
|
| 31 |
+
"encoder_config": {
|
| 32 |
+
"hidden_dim": 1024,
|
| 33 |
+
"ffn_dim": 4096,
|
| 34 |
+
"num_heads": 16,
|
| 35 |
+
"num_layers": 4
|
| 36 |
+
},
|
| 37 |
+
"dit_config": {
|
| 38 |
+
"hidden_dim": 1024,
|
| 39 |
+
"ffn_dim": 4096,
|
| 40 |
+
"num_heads": 16,
|
| 41 |
+
"num_layers": 4,
|
| 42 |
+
"cfm_config": {
|
| 43 |
+
"sigma_min": 1e-06,
|
| 44 |
+
"solver": "euler",
|
| 45 |
+
"t_scheduler": "log-norm",
|
| 46 |
+
"inference_cfg_rate": 2.0
|
| 47 |
+
}
|
| 48 |
+
},
|
| 49 |
+
"max_length": 4096,
|
| 50 |
+
"device": "cuda",
|
| 51 |
+
"dtype": "bfloat16"
|
| 52 |
+
}
|