add <|audio|> token
Browse files- config.json +79 -1
config.json
CHANGED
|
@@ -1 +1,79 @@
|
|
| 1 |
-
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/Users/zhuang/repos/ultravox-omni/artifacts/model-zhuang.2025-01-08-v0_5.llama3_2-1b-4a.246352f:v8",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"UltravoxModel"
|
| 5 |
+
],
|
| 6 |
+
"audio_config": {
|
| 7 |
+
"_name_or_path": "openai/whisper-large-v3-turbo",
|
| 8 |
+
"activation_dropout": 0.0,
|
| 9 |
+
"activation_function": "gelu",
|
| 10 |
+
"apply_spec_augment": false,
|
| 11 |
+
"architectures": [
|
| 12 |
+
"WhisperForConditionalGeneration"
|
| 13 |
+
],
|
| 14 |
+
"attention_dropout": 0.0,
|
| 15 |
+
"begin_suppress_tokens": [
|
| 16 |
+
220,
|
| 17 |
+
50256
|
| 18 |
+
],
|
| 19 |
+
"bos_token_id": 50257,
|
| 20 |
+
"d_model": 1280,
|
| 21 |
+
"decoder_attention_heads": 20,
|
| 22 |
+
"decoder_ffn_dim": 5120,
|
| 23 |
+
"decoder_layerdrop": 0.0,
|
| 24 |
+
"decoder_layers": 4,
|
| 25 |
+
"decoder_start_token_id": 50258,
|
| 26 |
+
"dropout": 0.0,
|
| 27 |
+
"encoder_attention_heads": 20,
|
| 28 |
+
"encoder_ffn_dim": 5120,
|
| 29 |
+
"encoder_layerdrop": 0.0,
|
| 30 |
+
"encoder_layers": 32,
|
| 31 |
+
"eos_token_id": 50257,
|
| 32 |
+
"init_std": 0.02,
|
| 33 |
+
"is_encoder_decoder": true,
|
| 34 |
+
"max_source_positions": 1500,
|
| 35 |
+
"max_target_positions": 448,
|
| 36 |
+
"median_filter_width": 7,
|
| 37 |
+
"model_type": "whisper",
|
| 38 |
+
"num_hidden_layers": 32,
|
| 39 |
+
"num_mel_bins": 128,
|
| 40 |
+
"pad_token_id": 50257,
|
| 41 |
+
"scale_embedding": false,
|
| 42 |
+
"torch_dtype": "float16",
|
| 43 |
+
"use_cache": true,
|
| 44 |
+
"vocab_size": 51866
|
| 45 |
+
},
|
| 46 |
+
"audio_latency_block_size": null,
|
| 47 |
+
"audio_model_id": null,
|
| 48 |
+
"auto_map": {
|
| 49 |
+
"AutoConfig": "ultravox_config.UltravoxConfig",
|
| 50 |
+
"AutoModel": "ultravox_model.UltravoxModel",
|
| 51 |
+
"AutoProcessor": "ultravox_processing.UltravoxProcessor"
|
| 52 |
+
},
|
| 53 |
+
"custom_pipelines": {
|
| 54 |
+
"ultravox-pipeline": {
|
| 55 |
+
"impl": "ultravox_pipeline.UltravoxPipeline",
|
| 56 |
+
"pt": [
|
| 57 |
+
"AutoModel"
|
| 58 |
+
],
|
| 59 |
+
"tf": [],
|
| 60 |
+
"type": "multimodal"
|
| 61 |
+
}
|
| 62 |
+
},
|
| 63 |
+
"hidden_size": 4096,
|
| 64 |
+
"ignore_index": -100,
|
| 65 |
+
"initializer_range": 0.02,
|
| 66 |
+
"model_type": "ultravox",
|
| 67 |
+
"norm_init": 0.4,
|
| 68 |
+
"pad_token_id": 128009,
|
| 69 |
+
"projector_act": "swiglu",
|
| 70 |
+
"projector_ln_mid": true,
|
| 71 |
+
"stack_factor": 8,
|
| 72 |
+
"text_model_id": "meta-llama/Llama-3.2-1B-Instruct",
|
| 73 |
+
"torch_dtype": "bfloat16",
|
| 74 |
+
"transformers_version": "4.48.1",
|
| 75 |
+
"num_attention_heads": 32,
|
| 76 |
+
"num_hidden_layers": 16,
|
| 77 |
+
"vocab_size": 128256,
|
| 78 |
+
"audio_token_index": 128256
|
| 79 |
+
}
|