model/config.json

{
  "_name_or_path": "ThivyanRR/indic_seamlessm4t_v2_large",
  "activation_dropout": 0.0,
  "activation_function": "relu",
  "adaptor_dropout": 0.1,
  "adaptor_kernel_size": 8,
  "adaptor_stride": 8,
  "add_adapter": true,
  "architectures": [
    "SeamlessM4Tv2Model"
  ],
  "attention_dropout": 0.1,
  "bos_token_id": 2,
  "char_vocab_size": 10943,
  "conv_depthwise_kernel_size": 31,
  "decoder_attention_heads": 16,
  "decoder_ffn_dim": 8192,
  "decoder_layerdrop": 0.05,
  "decoder_layers": 24,
  "decoder_start_token_id": 3,
  "dropout": 0.1,
  "encoder_attention_heads": 16,
  "encoder_ffn_dim": 8192,
  "encoder_layerdrop": 0.05,
  "encoder_layers": 24,
  "eos_token_id": 3,
  "feature_projection_input_dim": 160,
  "hidden_size": 1024,
  "initializer_range": 0.02,
  "is_encoder_decoder": true,
  "lang_embed_dim": 256,
  "layer_norm_eps": 1e-05,
  "leaky_relu_slope": 0.1,
  "left_max_position_embeddings": 64,
  "max_new_tokens": 256,
  "max_position_embeddings": 4096,
  "model_type": "seamless_m4t_v2",
  "num_adapter_layers": 1,
  "num_attention_heads": 16,
  "num_hidden_layers": 24,
  "pad_token_id": 0,
  "position_embeddings_type": "relative_key",
  "resblock_dilation_sizes": [
    [
      1,
      3,
      5
    ],
    [
      1,
      3,
      5
    ],
    [
      1,
      3,
      5
    ]
  ],
  "resblock_kernel_sizes": [
    3,
    7,
    11
  ],
  "right_max_position_embeddings": 8,
  "sampling_rate": 16000,
  "scale_embedding": true,
  "speech_encoder_attention_heads": 16,
  "speech_encoder_chunk_size": 20000,
  "speech_encoder_dropout": 0.0,
  "speech_encoder_hidden_act": "swish",
  "speech_encoder_intermediate_size": 4096,
  "speech_encoder_layerdrop": 0.1,
  "speech_encoder_layers": 24,
  "speech_encoder_left_chunk_num": 128,
  "spkr_embed_dim": 256,
  "t2u_bos_token_id": 0,
  "t2u_decoder_attention_heads": 16,
  "t2u_decoder_ffn_dim": 8192,
  "t2u_decoder_layers": 6,
  "t2u_encoder_attention_heads": 16,
  "t2u_encoder_ffn_dim": 8192,
  "t2u_encoder_layers": 6,
  "t2u_eos_token_id": 2,
  "t2u_max_position_embeddings": 4096,
  "t2u_pad_token_id": 1,
  "t2u_variance_pred_dropout": 0.5,
  "t2u_variance_predictor_embed_dim": 1024,
  "t2u_variance_predictor_hidden_dim": 256,
  "t2u_variance_predictor_kernel_size": 3,
  "t2u_vocab_size": 10082,
  "torch_dtype": "float32",
  "transformers_version": "4.45.1",
  "unit_embed_dim": 1280,
  "unit_hifi_gan_vocab_size": 10000,
  "upsample_initial_channel": 512,
  "upsample_kernel_sizes": [
    11,
    8,
    8,
    4,
    4
  ],
  "upsample_rates": [
    5,
    4,
    4,
    2,
    2
  ],
  "use_cache": true,
  "var_pred_dropout": 0.5,
  "variance_predictor_kernel_size": 3,
  "vocab_size": 256102,
  "vocoder_num_langs": 42,
  "vocoder_num_spkrs": 200,
  "vocoder_offset": 4,
  "weight_map": {
    "model.text_decoder_frontend.embed.weight": "model_shared.safetensors",
    "model.speech_encoder_frontend.post_extract_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder_frontend.post_extract_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder_frontend.model_dim_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder_frontend.model_dim_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.0.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.1.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.2.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.3.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.4.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.5.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.6.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.7.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.8.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.9.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.10.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.11.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.12.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.13.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.14.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.15.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.16.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.17.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.18.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.19.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.20.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.21.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.22.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.ffn1_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.ffn1_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.ffn1.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.ffn1.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.ffn1.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.ffn1.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn.sdpa.u_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn.sdpa.v_bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn.sdpa.r_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.self_attn.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.conv_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.conv_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.conv.pointwise_conv1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.conv.depthwise_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.conv.batch_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.conv.batch_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.conv.pointwise_conv2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.ffn2_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.ffn2_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.ffn2.inner_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.ffn2.inner_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.ffn2.output_proj.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.ffn2.output_proj.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner.layers.23.layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.inner_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.proj1.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.proj1.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.proj2.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.proj2.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.residual_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.residual_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.residual_conv.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.residual_conv.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.self_attn_conv.weight": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.self_attn_conv.bias": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.speech_encoder.adaptor_layers.0.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.speech_encoder.layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.speech_encoder.layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_encoder_frontend.embed.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.0.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.1.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.2.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.3.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.4.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.5.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.6.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.7.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.8.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.9.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.10.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.11.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.12.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.13.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.14.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.15.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.16.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.17.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.18.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.19.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.20.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.21.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.22.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layers.23.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.text_decoder.layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.text_decoder.layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.final_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.0.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.1.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.2.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.3.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.4.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layers.5.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.encoder.layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder_frontend.embed.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.0.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.1.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.2.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.3.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.4.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.self_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.self_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.self_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.self_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.encoder_decoder_attn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.encoder_decoder_attn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.encoder_decoder_attn.q_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.encoder_decoder_attn.q_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.encoder_decoder_attn.k_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.encoder_decoder_attn.k_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.encoder_decoder_attn.v_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.encoder_decoder_attn.v_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.encoder_decoder_attn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.encoder_decoder_attn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.ffn_layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.ffn_layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.ffn.inner_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.ffn.inner_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.ffn.output_proj.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layers.5.ffn.output_proj.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layer_norm.weight": "model-00002-of-00002.safetensors",
    "model.t2u_model.decoder.layer_norm.bias": "model-00002-of-00002.safetensors",
    "model.t2u_model.final_proj.weight": "model-00002-of-00002.safetensors"
  }
}