Blinorot/AL-SSLAM-R

1{2  "architectures": [3    "Qwen3AudioWrappedForCausalLM"4  ],5  "attention_bias": false,6  "attention_dropout": 0.0,7  "audio_adapter_configs": [8    {9      "adapter_embedding_dim": 1024,10      "adapter_name": "downsampler_conformer",11      "audio_encoder_layers": [12        3,13        7,14        1115      ],16      "downsampler_depth": 1,17      "encoder_embedding_dim": 768,18      "encoder_name": "sslam",19      "layer_fusion_config": {20        "layer_fusion_type": "weighted_average"21      },22      "llm_embedding_dim": 2560,23      "norm_type": "layer",24      "pre_average": true,25      "use_conformer": false26    }27  ],28  "audio_encoder_configs": [29    {30      "encoder_name": "sslam"31    }32  ],33  "audio_fusion_config": {34    "fusion_type": "sum"35  },36  "audio_postprocessing_config": {37    "postprocessing_type": "identity"38  },39  "audio_sep_d_embed": 2560,40  "bos_token_id": 151643,41  "dtype": "float32",42  "eos_token_id": 151645,43  "head_dim": 128,44  "hidden_act": "silu",45  "hidden_size": 2560,46  "initializer_range": 0.02,47  "intermediate_size": 9728,48  "layer_types": [49    "full_attention",50    "full_attention",51    "full_attention",52    "full_attention",53    "full_attention",54    "full_attention",55    "full_attention",56    "full_attention",57    "full_attention",58    "full_attention",59    "full_attention",60    "full_attention",61    "full_attention",62    "full_attention",63    "full_attention",64    "full_attention",65    "full_attention",66    "full_attention",67    "full_attention",68    "full_attention",69    "full_attention",70    "full_attention",71    "full_attention",72    "full_attention",73    "full_attention",74    "full_attention",75    "full_attention",76    "full_attention",77    "full_attention",78    "full_attention",79    "full_attention",80    "full_attention",81    "full_attention",82    "full_attention",83    "full_attention",84    "full_attention"85  ],86  "max_position_embeddings": 262144,87  "max_window_layers": 36,88  "model_type": "qwen3_audio",89  "num_attention_heads": 32,90  "num_hidden_layers": 36,91  "num_key_value_heads": 8,92  "rms_norm_eps": 1e-06,93  "rope_scaling": null,94  "rope_theta": 5000000,95  "sliding_window": null,96  "tie_word_embeddings": true,97  "transformers_version": "4.57.1",98  "use_cache": false,99  "use_explicit_audio_tokens": false,100  "use_sliding_window": false,101  "vocab_size": 151936102}103