audio-flamingo-3-hf / config.json
SreyanG-NVIDIA's picture
Change init_std to initializer_range
8d47efd verified
raw
history blame
2.12 kB
{
"architectures": [
"AudioFlamingo3ForConditionalGeneration"
],
"audio_config": {
"activation_dropout": 0.0,
"activation_function": "gelu",
"attention_dropout": 0.0,
"dropout": 0.0,
"hidden_size": 1280,
"initializer_range": 0.02,
"intermediate_size": 5120,
"layerdrop": 0.0,
"max_source_positions": 1500,
"model_type": "audioflamingo3_encoder",
"num_attention_heads": 20,
"num_hidden_layers": 32,
"num_mel_bins": 128,
"scale_embedding": false
},
"audio_token_id": 151669,
"dtype": "float32",
"model_type": "audioflamingo3",
"projector_bias": true,
"projector_hidden_act": "gelu",
"text_config": {
"attention_dropout": 0.0,
"bos_token_id": 151643,
"dtype": "bfloat16",
"eos_token_id": 151645,
"hidden_act": "silu",
"hidden_size": 3584,
"initializer_range": 0.02,
"intermediate_size": 18944,
"layer_types": [
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention",
"full_attention"
],
"max_position_embeddings": 32768,
"max_window_layers": 28,
"model_max_length": 8192,
"model_type": "qwen2",
"num_attention_heads": 28,
"num_hidden_layers": 28,
"num_key_value_heads": 4,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"rope_theta": 1000000.0,
"rope_type": "default"
},
"rope_theta": 1000000.0,
"sliding_window": null,
"use_cache": false,
"use_sliding_window": false,
"vocab_size": 151672
},
"transformers_version": "5.0.0.dev0"
}