diff --git "a/model.safetensors.index.json" "b/model.safetensors.index.json" new file mode 100644--- /dev/null +++ "b/model.safetensors.index.json" @@ -0,0 +1,1212 @@ +{ + "metadata": { + "total_parameters": 2704021987, + "total_size": 5408043974 + }, + "weight_map": { + "latent_bias_factor": "model-00001-of-00003.safetensors", + "latent_scaling_factor": "model-00001-of-00003.safetensors", + "model.acoustic_connector.fc1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_connector.fc1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_connector.fc2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_connector.fc2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_connector.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.convtr.convtr.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.convtr.convtr.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.0.stage.2.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.convtr.convtr.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.convtr.convtr.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.1.stage.2.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.convtr.convtr.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.convtr.convtr.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.2.stage.2.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.convtr.convtr.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.convtr.convtr.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.3.stage.2.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.convtr.convtr.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.convtr.convtr.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.4.stage.2.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.convtr.convtr.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.convtr.convtr.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.conv_layers.5.stage.2.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.head.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.head.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.conv.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.conv.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.0.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.0.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.0.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.0.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.0.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.0.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.0.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.0.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.0.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.0.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.1.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.1.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.1.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.1.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.1.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.1.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.1.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.1.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.1.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.1.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.2.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.2.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.2.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.2.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.2.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.2.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.3.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.3.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.3.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.3.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.3.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.3.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.3.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.3.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.3.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.3.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.4.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.4.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.4.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.4.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.4.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.4.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.4.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.4.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.4.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.4.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.5.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.5.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.5.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.5.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.5.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.5.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.5.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.5.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.5.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.5.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.6.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.6.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.6.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.6.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.6.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.6.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.6.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.6.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.6.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.6.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.7.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.7.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.7.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.7.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.7.ffn_gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.7.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.7.gamma": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.7.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.7.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.decoder.stem.stage.7.norm.weight": "model-00003-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.conv.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.conv.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.0.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.0.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.0.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.0.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.0.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.0.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.0.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.0.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.0.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.0.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.1.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.1.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.1.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.1.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.1.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.1.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.1.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.1.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.1.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.1.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.2.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.2.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.2.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.2.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.2.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.2.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.2.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.2.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.2.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.0.stage.2.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.conv.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.conv.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.0.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.0.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.0.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.0.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.0.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.0.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.0.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.0.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.0.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.0.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.1.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.1.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.1.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.1.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.1.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.1.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.1.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.1.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.1.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.1.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.2.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.2.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.2.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.2.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.2.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.2.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.2.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.2.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.2.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.1.stage.2.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.conv.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.conv.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.0.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.0.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.0.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.0.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.0.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.0.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.0.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.0.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.0.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.0.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.1.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.1.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.1.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.1.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.1.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.1.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.1.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.1.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.1.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.1.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.2.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.2.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.2.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.2.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.2.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.2.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.2.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.2.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.2.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.2.stage.2.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.conv.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.conv.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.0.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.0.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.0.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.0.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.0.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.0.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.0.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.0.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.0.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.0.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.1.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.1.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.1.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.1.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.1.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.1.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.1.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.1.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.1.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.1.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.2.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.2.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.2.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.2.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.2.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.2.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.2.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.2.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.2.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.3.stage.2.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.conv.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.conv.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.0.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.0.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.0.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.0.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.0.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.0.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.0.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.0.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.0.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.0.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.1.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.1.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.1.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.1.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.1.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.1.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.1.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.1.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.1.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.1.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.2.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.2.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.2.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.2.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.2.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.2.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.2.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.2.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.2.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.4.stage.2.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.conv.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.conv.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.0.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.0.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.0.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.0.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.0.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.0.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.0.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.0.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.0.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.0.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.1.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.1.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.1.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.1.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.1.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.1.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.1.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.1.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.1.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.1.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.2.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.2.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.2.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.2.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.2.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.2.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.2.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.2.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.2.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.2.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.3.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.3.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.3.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.3.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.3.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.3.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.3.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.3.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.3.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.3.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.4.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.4.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.4.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.4.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.4.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.4.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.4.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.4.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.4.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.4.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.5.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.5.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.5.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.5.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.5.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.5.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.5.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.5.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.5.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.5.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.6.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.6.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.6.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.6.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.6.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.6.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.6.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.6.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.6.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.6.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.7.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.7.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.7.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.7.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.7.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.7.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.7.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.7.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.7.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.conv_layers.5.stage.7.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.head.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.head.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.conv.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.conv.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.0.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.0.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.0.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.0.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.0.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.0.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.0.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.0.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.0.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.0.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.1.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.1.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.1.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.1.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.1.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.1.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.1.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.1.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.1.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.1.norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.2.ffn.linear1.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.2.ffn.linear1.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.2.ffn.linear2.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.2.ffn.linear2.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.2.ffn_gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.2.ffn_norm.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.2.gamma": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.2.mixer.conv.bias": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.2.mixer.conv.weight": "model-00002-of-00003.safetensors", + "model.acoustic_tokenizer.encoder.stem.stage.2.norm.weight": "model-00002-of-00003.safetensors", + "model.diffusion_head.cond_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.final_layer.linear_1.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.final_layer.linear_2.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.0.ffn.down_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.0.ffn.gate_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.0.ffn.up_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.0.linear.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.0.norm.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.1.ffn.down_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.1.ffn.gate_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.1.ffn.up_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.1.linear.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.1.norm.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.2.ffn.down_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.2.ffn.gate_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.2.ffn.up_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.2.linear.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.2.norm.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.3.ffn.down_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.3.ffn.gate_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.3.ffn.up_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.3.linear.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.layers.3.norm.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.noisy_images_proj.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.timestep_embedder.layer_1.weight": "model-00003-of-00003.safetensors", + "model.diffusion_head.timestep_embedder.layer_2.weight": "model-00003-of-00003.safetensors", + "model.language_model.embed_tokens.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.0.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.0.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.0.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.1.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.1.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.1.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.10.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.10.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.10.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.11.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.11.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.11.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.11.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.12.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.12.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.12.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.13.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.13.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.13.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.14.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.14.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.14.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.15.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.15.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.15.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.16.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.16.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.16.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.17.self_attn.k_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.17.self_attn.q_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.17.self_attn.v_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.18.self_attn.k_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.18.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.18.self_attn.q_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.18.self_attn.v_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.19.self_attn.k_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.19.self_attn.q_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.19.self_attn.v_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.2.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.2.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.2.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.20.self_attn.k_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.20.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.20.self_attn.q_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.20.self_attn.v_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.21.self_attn.k_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.21.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.21.self_attn.q_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.21.self_attn.v_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.22.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.22.mlp.down_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.22.mlp.gate_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.22.mlp.up_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.22.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.22.self_attn.k_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.22.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.22.self_attn.q_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.22.self_attn.v_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.23.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.23.mlp.down_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.23.mlp.gate_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.23.mlp.up_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.23.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.23.self_attn.k_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.23.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.23.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.23.self_attn.q_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.23.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.23.self_attn.v_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.23.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.24.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.24.mlp.down_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.24.mlp.gate_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.24.mlp.up_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.24.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.24.self_attn.k_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.24.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.24.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.24.self_attn.q_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.24.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.24.self_attn.v_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.24.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.25.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.25.mlp.down_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.25.mlp.gate_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.25.mlp.up_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.25.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.25.self_attn.k_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.25.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.25.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.25.self_attn.q_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.25.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.25.self_attn.v_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.25.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.26.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.26.mlp.down_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.26.mlp.gate_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.26.mlp.up_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.26.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.26.self_attn.k_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.26.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.26.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.26.self_attn.q_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.26.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.26.self_attn.v_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.26.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.27.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.27.mlp.down_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.27.mlp.gate_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.27.mlp.up_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.27.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.27.self_attn.k_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.27.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.27.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.27.self_attn.q_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.27.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.27.self_attn.v_proj.bias": "model-00002-of-00003.safetensors", + "model.language_model.layers.27.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.3.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.3.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.3.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.4.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.4.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.4.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.5.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.5.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.5.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.6.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.6.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.6.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.7.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.7.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.7.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.8.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.8.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.8.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.9.self_attn.k_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.9.self_attn.q_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.layers.9.self_attn.v_proj.bias": "model-00001-of-00003.safetensors", + "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.language_model.norm.weight": "model-00002-of-00003.safetensors", + "model.semantic_connector.fc1.bias": "model-00003-of-00003.safetensors", + "model.semantic_connector.fc1.weight": "model-00003-of-00003.safetensors", + "model.semantic_connector.fc2.bias": "model-00003-of-00003.safetensors", + "model.semantic_connector.fc2.weight": "model-00003-of-00003.safetensors", + "model.semantic_connector.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.conv.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.conv.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.0.stage.2.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.conv.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.conv.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.1.stage.2.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.conv.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.conv.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.2.stage.2.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.conv.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.conv.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.3.stage.2.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.conv.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.conv.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.4.stage.2.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.conv.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.conv.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.2.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.3.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.3.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.3.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.3.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.3.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.3.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.3.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.3.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.3.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.3.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.4.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.4.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.4.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.4.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.4.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.4.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.4.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.4.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.4.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.4.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.5.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.5.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.5.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.5.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.5.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.5.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.5.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.5.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.5.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.5.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.6.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.6.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.6.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.6.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.6.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.6.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.6.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.6.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.6.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.6.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.7.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.7.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.7.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.7.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.7.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.7.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.7.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.7.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.7.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.conv_layers.5.stage.7.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.head.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.head.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.conv.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.conv.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.0.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.0.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.0.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.0.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.0.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.0.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.0.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.0.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.0.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.0.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.1.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.1.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.1.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.1.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.1.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.1.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.1.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.1.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.1.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.1.norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.2.ffn.linear1.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.2.ffn.linear1.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.2.ffn.linear2.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.2.ffn.linear2.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.2.ffn_gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.2.ffn_norm.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.2.gamma": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.2.mixer.conv.bias": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.2.mixer.conv.weight": "model-00003-of-00003.safetensors", + "model.semantic_tokenizer.encoder.stem.stage.2.norm.weight": "model-00003-of-00003.safetensors" + } +}