[2026-09-06T09:30:37Z] EMBED 2241 crops to embed music/venvs/eval/.venv/lib/python3.12/site-packages/torch/functional.py:554: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /pytorch/aten/src/ATen/native/TensorShape.cpp:4322.) return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] Some weights of RobertaModel were not initialized from the model checkpoint at roberta-base and are newly initialized: ['pooler.dense.bias', 'pooler.dense.weight'] You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference. Load the specified checkpoint music/venvs/eval/weights/music_audioset_epoch_15_esc_90.14.pt from users. Load Checkpoint... logit_scale_a Loaded logit_scale_t Loaded audio_branch.spectrogram_extractor.stft.conv_real.weight Loaded audio_branch.spectrogram_extractor.stft.conv_imag.weight Loaded audio_branch.logmel_extractor.melW Loaded audio_branch.bn0.weight Loaded audio_branch.bn0.bias Loaded audio_branch.patch_embed.proj.weight Loaded audio_branch.patch_embed.proj.bias Loaded audio_branch.patch_embed.norm.weight Loaded audio_branch.patch_embed.norm.bias Loaded audio_branch.layers.0.blocks.0.norm1.weight Loaded audio_branch.layers.0.blocks.0.norm1.bias Loaded audio_branch.layers.0.blocks.0.attn.relative_position_bias_table Loaded audio_branch.layers.0.blocks.0.attn.qkv.weight Loaded audio_branch.layers.0.blocks.0.attn.qkv.bias Loaded audio_branch.layers.0.blocks.0.attn.proj.weight Loaded audio_branch.layers.0.blocks.0.attn.proj.bias Loaded audio_branch.layers.0.blocks.0.norm2.weight Loaded audio_branch.layers.0.blocks.0.norm2.bias Loaded audio_branch.layers.0.blocks.0.mlp.fc1.weight Loaded audio_branch.layers.0.blocks.0.mlp.fc1.bias Loaded audio_branch.layers.0.blocks.0.mlp.fc2.weight Loaded audio_branch.layers.0.blocks.0.mlp.fc2.bias Loaded audio_branch.layers.0.blocks.1.norm1.weight Loaded audio_branch.layers.0.blocks.1.norm1.bias Loaded audio_branch.layers.0.blocks.1.attn.relative_position_bias_table Loaded audio_branch.layers.0.blocks.1.attn.qkv.weight Loaded audio_branch.layers.0.blocks.1.attn.qkv.bias Loaded audio_branch.layers.0.blocks.1.attn.proj.weight Loaded audio_branch.layers.0.blocks.1.attn.proj.bias Loaded audio_branch.layers.0.blocks.1.norm2.weight Loaded audio_branch.layers.0.blocks.1.norm2.bias Loaded audio_branch.layers.0.blocks.1.mlp.fc1.weight Loaded audio_branch.layers.0.blocks.1.mlp.fc1.bias Loaded audio_branch.layers.0.blocks.1.mlp.fc2.weight Loaded audio_branch.layers.0.blocks.1.mlp.fc2.bias Loaded audio_branch.layers.0.downsample.reduction.weight Loaded audio_branch.layers.0.downsample.norm.weight Loaded audio_branch.layers.0.downsample.norm.bias Loaded audio_branch.layers.1.blocks.0.norm1.weight Loaded audio_branch.layers.1.blocks.0.norm1.bias Loaded audio_branch.layers.1.blocks.0.attn.relative_position_bias_table Loaded audio_branch.layers.1.blocks.0.attn.qkv.weight Loaded audio_branch.layers.1.blocks.0.attn.qkv.bias Loaded audio_branch.layers.1.blocks.0.attn.proj.weight Loaded audio_branch.layers.1.blocks.0.attn.proj.bias Loaded audio_branch.layers.1.blocks.0.norm2.weight Loaded audio_branch.layers.1.blocks.0.norm2.bias Loaded audio_branch.layers.1.blocks.0.mlp.fc1.weight Loaded audio_branch.layers.1.blocks.0.mlp.fc1.bias Loaded audio_branch.layers.1.blocks.0.mlp.fc2.weight Loaded audio_branch.layers.1.blocks.0.mlp.fc2.bias Loaded audio_branch.layers.1.blocks.1.norm1.weight Loaded audio_branch.layers.1.blocks.1.norm1.bias Loaded audio_branch.layers.1.blocks.1.attn.relative_position_bias_table Loaded audio_branch.layers.1.blocks.1.attn.qkv.weight Loaded audio_branch.layers.1.blocks.1.attn.qkv.bias Loaded audio_branch.layers.1.blocks.1.attn.proj.weight Loaded audio_branch.layers.1.blocks.1.attn.proj.bias Loaded audio_branch.layers.1.blocks.1.norm2.weight Loaded audio_branch.layers.1.blocks.1.norm2.bias Loaded audio_branch.layers.1.blocks.1.mlp.fc1.weight Loaded audio_branch.layers.1.blocks.1.mlp.fc1.bias Loaded audio_branch.layers.1.blocks.1.mlp.fc2.weight Loaded audio_branch.layers.1.blocks.1.mlp.fc2.bias Loaded audio_branch.layers.1.downsample.reduction.weight Loaded audio_branch.layers.1.downsample.norm.weight Loaded audio_branch.layers.1.downsample.norm.bias Loaded audio_branch.layers.2.blocks.0.norm1.weight Loaded audio_branch.layers.2.blocks.0.norm1.bias Loaded audio_branch.layers.2.blocks.0.attn.relative_position_bias_table Loaded audio_branch.layers.2.blocks.0.attn.qkv.weight Loaded audio_branch.layers.2.blocks.0.attn.qkv.bias Loaded audio_branch.layers.2.blocks.0.attn.proj.weight Loaded audio_branch.layers.2.blocks.0.attn.proj.bias Loaded audio_branch.layers.2.blocks.0.norm2.weight Loaded audio_branch.layers.2.blocks.0.norm2.bias Loaded audio_branch.layers.2.blocks.0.mlp.fc1.weight Loaded audio_branch.layers.2.blocks.0.mlp.fc1.bias Loaded audio_branch.layers.2.blocks.0.mlp.fc2.weight Loaded audio_branch.layers.2.blocks.0.mlp.fc2.bias Loaded audio_branch.layers.2.blocks.1.norm1.weight Loaded audio_branch.layers.2.blocks.1.norm1.bias Loaded audio_branch.layers.2.blocks.1.attn.relative_position_bias_table Loaded audio_branch.layers.2.blocks.1.attn.qkv.weight Loaded audio_branch.layers.2.blocks.1.attn.qkv.bias Loaded audio_branch.layers.2.blocks.1.attn.proj.weight Loaded audio_branch.layers.2.blocks.1.attn.proj.bias Loaded audio_branch.layers.2.blocks.1.norm2.weight Loaded audio_branch.layers.2.blocks.1.norm2.bias Loaded audio_branch.layers.2.blocks.1.mlp.fc1.weight Loaded audio_branch.layers.2.blocks.1.mlp.fc1.bias Loaded audio_branch.layers.2.blocks.1.mlp.fc2.weight Loaded audio_branch.layers.2.blocks.1.mlp.fc2.bias Loaded audio_branch.layers.2.blocks.2.norm1.weight Loaded audio_branch.layers.2.blocks.2.norm1.bias Loaded audio_branch.layers.2.blocks.2.attn.relative_position_bias_table Loaded audio_branch.layers.2.blocks.2.attn.qkv.weight Loaded audio_branch.layers.2.blocks.2.attn.qkv.bias Loaded audio_branch.layers.2.blocks.2.attn.proj.weight Loaded audio_branch.layers.2.blocks.2.attn.proj.bias Loaded audio_branch.layers.2.blocks.2.norm2.weight Loaded audio_branch.layers.2.blocks.2.norm2.bias Loaded audio_branch.layers.2.blocks.2.mlp.fc1.weight Loaded audio_branch.layers.2.blocks.2.mlp.fc1.bias Loaded audio_branch.layers.2.blocks.2.mlp.fc2.weight Loaded audio_branch.layers.2.blocks.2.mlp.fc2.bias Loaded audio_branch.layers.2.blocks.3.norm1.weight Loaded audio_branch.layers.2.blocks.3.norm1.bias Loaded audio_branch.layers.2.blocks.3.attn.relative_position_bias_table Loaded audio_branch.layers.2.blocks.3.attn.qkv.weight Loaded audio_branch.layers.2.blocks.3.attn.qkv.bias Loaded audio_branch.layers.2.blocks.3.attn.proj.weight Loaded audio_branch.layers.2.blocks.3.attn.proj.bias Loaded audio_branch.layers.2.blocks.3.norm2.weight Loaded audio_branch.layers.2.blocks.3.norm2.bias Loaded audio_branch.layers.2.blocks.3.mlp.fc1.weight Loaded audio_branch.layers.2.blocks.3.mlp.fc1.bias Loaded audio_branch.layers.2.blocks.3.mlp.fc2.weight Loaded audio_branch.layers.2.blocks.3.mlp.fc2.bias Loaded audio_branch.layers.2.blocks.4.norm1.weight Loaded audio_branch.layers.2.blocks.4.norm1.bias Loaded audio_branch.layers.2.blocks.4.attn.relative_position_bias_table Loaded audio_branch.layers.2.blocks.4.attn.qkv.weight Loaded audio_branch.layers.2.blocks.4.attn.qkv.bias Loaded audio_branch.layers.2.blocks.4.attn.proj.weight Loaded audio_branch.layers.2.blocks.4.attn.proj.bias Loaded audio_branch.layers.2.blocks.4.norm2.weight Loaded audio_branch.layers.2.blocks.4.norm2.bias Loaded audio_branch.layers.2.blocks.4.mlp.fc1.weight Loaded audio_branch.layers.2.blocks.4.mlp.fc1.bias Loaded audio_branch.layers.2.blocks.4.mlp.fc2.weight Loaded audio_branch.layers.2.blocks.4.mlp.fc2.bias Loaded audio_branch.layers.2.blocks.5.norm1.weight Loaded audio_branch.layers.2.blocks.5.norm1.bias Loaded audio_branch.layers.2.blocks.5.attn.relative_position_bias_table Loaded audio_branch.layers.2.blocks.5.attn.qkv.weight Loaded audio_branch.layers.2.blocks.5.attn.qkv.bias Loaded audio_branch.layers.2.blocks.5.attn.proj.weight Loaded audio_branch.layers.2.blocks.5.attn.proj.bias Loaded audio_branch.layers.2.blocks.5.norm2.weight Loaded audio_branch.layers.2.blocks.5.norm2.bias Loaded audio_branch.layers.2.blocks.5.mlp.fc1.weight Loaded audio_branch.layers.2.blocks.5.mlp.fc1.bias Loaded audio_branch.layers.2.blocks.5.mlp.fc2.weight Loaded audio_branch.layers.2.blocks.5.mlp.fc2.bias Loaded audio_branch.layers.2.blocks.6.norm1.weight Loaded audio_branch.layers.2.blocks.6.norm1.bias Loaded audio_branch.layers.2.blocks.6.attn.relative_position_bias_table Loaded audio_branch.layers.2.blocks.6.attn.qkv.weight Loaded audio_branch.layers.2.blocks.6.attn.qkv.bias Loaded audio_branch.layers.2.blocks.6.attn.proj.weight Loaded audio_branch.layers.2.blocks.6.attn.proj.bias Loaded audio_branch.layers.2.blocks.6.norm2.weight Loaded audio_branch.layers.2.blocks.6.norm2.bias Loaded audio_branch.layers.2.blocks.6.mlp.fc1.weight Loaded audio_branch.layers.2.blocks.6.mlp.fc1.bias Loaded audio_branch.layers.2.blocks.6.mlp.fc2.weight Loaded audio_branch.layers.2.blocks.6.mlp.fc2.bias Loaded audio_branch.layers.2.blocks.7.norm1.weight Loaded audio_branch.layers.2.blocks.7.norm1.bias Loaded audio_branch.layers.2.blocks.7.attn.relative_position_bias_table Loaded audio_branch.layers.2.blocks.7.attn.qkv.weight Loaded audio_branch.layers.2.blocks.7.attn.qkv.bias Loaded audio_branch.layers.2.blocks.7.attn.proj.weight Loaded audio_branch.layers.2.blocks.7.attn.proj.bias Loaded audio_branch.layers.2.blocks.7.norm2.weight Loaded audio_branch.layers.2.blocks.7.norm2.bias Loaded audio_branch.layers.2.blocks.7.mlp.fc1.weight Loaded audio_branch.layers.2.blocks.7.mlp.fc1.bias Loaded audio_branch.layers.2.blocks.7.mlp.fc2.weight Loaded audio_branch.layers.2.blocks.7.mlp.fc2.bias Loaded audio_branch.layers.2.blocks.8.norm1.weight Loaded audio_branch.layers.2.blocks.8.norm1.bias Loaded audio_branch.layers.2.blocks.8.attn.relative_position_bias_table Loaded audio_branch.layers.2.blocks.8.attn.qkv.weight Loaded audio_branch.layers.2.blocks.8.attn.qkv.bias Loaded audio_branch.layers.2.blocks.8.attn.proj.weight Loaded audio_branch.layers.2.blocks.8.attn.proj.bias Loaded audio_branch.layers.2.blocks.8.norm2.weight Loaded audio_branch.layers.2.blocks.8.norm2.bias Loaded audio_branch.layers.2.blocks.8.mlp.fc1.weight Loaded audio_branch.layers.2.blocks.8.mlp.fc1.bias Loaded audio_branch.layers.2.blocks.8.mlp.fc2.weight Loaded audio_branch.layers.2.blocks.8.mlp.fc2.bias Loaded audio_branch.layers.2.blocks.9.norm1.weight Loaded audio_branch.layers.2.blocks.9.norm1.bias Loaded audio_branch.layers.2.blocks.9.attn.relative_position_bias_table Loaded audio_branch.layers.2.blocks.9.attn.qkv.weight Loaded audio_branch.layers.2.blocks.9.attn.qkv.bias Loaded audio_branch.layers.2.blocks.9.attn.proj.weight Loaded audio_branch.layers.2.blocks.9.attn.proj.bias Loaded audio_branch.layers.2.blocks.9.norm2.weight Loaded audio_branch.layers.2.blocks.9.norm2.bias Loaded audio_branch.layers.2.blocks.9.mlp.fc1.weight Loaded audio_branch.layers.2.blocks.9.mlp.fc1.bias Loaded audio_branch.layers.2.blocks.9.mlp.fc2.weight Loaded audio_branch.layers.2.blocks.9.mlp.fc2.bias Loaded audio_branch.layers.2.blocks.10.norm1.weight Loaded audio_branch.layers.2.blocks.10.norm1.bias Loaded audio_branch.layers.2.blocks.10.attn.relative_position_bias_table Loaded audio_branch.layers.2.blocks.10.attn.qkv.weight Loaded audio_branch.layers.2.blocks.10.attn.qkv.bias Loaded audio_branch.layers.2.blocks.10.attn.proj.weight Loaded audio_branch.layers.2.blocks.10.attn.proj.bias Loaded audio_branch.layers.2.blocks.10.norm2.weight Loaded audio_branch.layers.2.blocks.10.norm2.bias Loaded audio_branch.layers.2.blocks.10.mlp.fc1.weight Loaded audio_branch.layers.2.blocks.10.mlp.fc1.bias Loaded audio_branch.layers.2.blocks.10.mlp.fc2.weight Loaded audio_branch.layers.2.blocks.10.mlp.fc2.bias Loaded audio_branch.layers.2.blocks.11.norm1.weight Loaded audio_branch.layers.2.blocks.11.norm1.bias Loaded audio_branch.layers.2.blocks.11.attn.relative_position_bias_table Loaded audio_branch.layers.2.blocks.11.attn.qkv.weight Loaded audio_branch.layers.2.blocks.11.attn.qkv.bias Loaded audio_branch.layers.2.blocks.11.attn.proj.weight Loaded audio_branch.layers.2.blocks.11.attn.proj.bias Loaded audio_branch.layers.2.blocks.11.norm2.weight Loaded audio_branch.layers.2.blocks.11.norm2.bias Loaded audio_branch.layers.2.blocks.11.mlp.fc1.weight Loaded audio_branch.layers.2.blocks.11.mlp.fc1.bias Loaded audio_branch.layers.2.blocks.11.mlp.fc2.weight Loaded audio_branch.layers.2.blocks.11.mlp.fc2.bias Loaded audio_branch.layers.2.downsample.reduction.weight Loaded audio_branch.layers.2.downsample.norm.weight Loaded audio_branch.layers.2.downsample.norm.bias Loaded audio_branch.layers.3.blocks.0.norm1.weight Loaded audio_branch.layers.3.blocks.0.norm1.bias Loaded audio_branch.layers.3.blocks.0.attn.relative_position_bias_table Loaded audio_branch.layers.3.blocks.0.attn.qkv.weight Loaded audio_branch.layers.3.blocks.0.attn.qkv.bias Loaded audio_branch.layers.3.blocks.0.attn.proj.weight Loaded audio_branch.layers.3.blocks.0.attn.proj.bias Loaded audio_branch.layers.3.blocks.0.norm2.weight Loaded audio_branch.layers.3.blocks.0.norm2.bias Loaded audio_branch.layers.3.blocks.0.mlp.fc1.weight Loaded audio_branch.layers.3.blocks.0.mlp.fc1.bias Loaded audio_branch.layers.3.blocks.0.mlp.fc2.weight Loaded audio_branch.layers.3.blocks.0.mlp.fc2.bias Loaded audio_branch.layers.3.blocks.1.norm1.weight Loaded audio_branch.layers.3.blocks.1.norm1.bias Loaded audio_branch.layers.3.blocks.1.attn.relative_position_bias_table Loaded audio_branch.layers.3.blocks.1.attn.qkv.weight Loaded audio_branch.layers.3.blocks.1.attn.qkv.bias Loaded audio_branch.layers.3.blocks.1.attn.proj.weight Loaded audio_branch.layers.3.blocks.1.attn.proj.bias Loaded audio_branch.layers.3.blocks.1.norm2.weight Loaded audio_branch.layers.3.blocks.1.norm2.bias Loaded audio_branch.layers.3.blocks.1.mlp.fc1.weight Loaded audio_branch.layers.3.blocks.1.mlp.fc1.bias Loaded audio_branch.layers.3.blocks.1.mlp.fc2.weight Loaded audio_branch.layers.3.blocks.1.mlp.fc2.bias Loaded audio_branch.norm.weight Loaded audio_branch.norm.bias Loaded audio_branch.tscam_conv.weight Loaded audio_branch.tscam_conv.bias Loaded audio_branch.head.weight Loaded audio_branch.head.bias Loaded text_branch.embeddings.word_embeddings.weight Loaded text_branch.embeddings.position_embeddings.weight Loaded text_branch.embeddings.token_type_embeddings.weight Loaded text_branch.embeddings.LayerNorm.weight Loaded text_branch.embeddings.LayerNorm.bias Loaded text_branch.encoder.layer.0.attention.self.query.weight Loaded text_branch.encoder.layer.0.attention.self.query.bias Loaded text_branch.encoder.layer.0.attention.self.key.weight Loaded text_branch.encoder.layer.0.attention.self.key.bias Loaded text_branch.encoder.layer.0.attention.self.value.weight Loaded text_branch.encoder.layer.0.attention.self.value.bias Loaded text_branch.encoder.layer.0.attention.output.dense.weight Loaded text_branch.encoder.layer.0.attention.output.dense.bias Loaded text_branch.encoder.layer.0.attention.output.LayerNorm.weight Loaded text_branch.encoder.layer.0.attention.output.LayerNorm.bias Loaded text_branch.encoder.layer.0.intermediate.dense.weight Loaded text_branch.encoder.layer.0.intermediate.dense.bias Loaded text_branch.encoder.layer.0.output.dense.weight Loaded text_branch.encoder.layer.0.output.dense.bias Loaded text_branch.encoder.layer.0.output.LayerNorm.weight Loaded text_branch.encoder.layer.0.output.LayerNorm.bias Loaded text_branch.encoder.layer.1.attention.self.query.weight Loaded text_branch.encoder.layer.1.attention.self.query.bias Loaded text_branch.encoder.layer.1.attention.self.key.weight Loaded text_branch.encoder.layer.1.attention.self.key.bias Loaded text_branch.encoder.layer.1.attention.self.value.weight Loaded text_branch.encoder.layer.1.attention.self.value.bias Loaded text_branch.encoder.layer.1.attention.output.dense.weight Loaded text_branch.encoder.layer.1.attention.output.dense.bias Loaded text_branch.encoder.layer.1.attention.output.LayerNorm.weight Loaded text_branch.encoder.layer.1.attention.output.LayerNorm.bias Loaded text_branch.encoder.layer.1.intermediate.dense.weight Loaded text_branch.encoder.layer.1.intermediate.dense.bias Loaded text_branch.encoder.layer.1.output.dense.weight Loaded text_branch.encoder.layer.1.output.dense.bias Loaded text_branch.encoder.layer.1.output.LayerNorm.weight Loaded text_branch.encoder.layer.1.output.LayerNorm.bias Loaded text_branch.encoder.layer.2.attention.self.query.weight Loaded text_branch.encoder.layer.2.attention.self.query.bias Loaded text_branch.encoder.layer.2.attention.self.key.weight Loaded text_branch.encoder.layer.2.attention.self.key.bias Loaded text_branch.encoder.layer.2.attention.self.value.weight Loaded text_branch.encoder.layer.2.attention.self.value.bias Loaded text_branch.encoder.layer.2.attention.output.dense.weight Loaded text_branch.encoder.layer.2.attention.output.dense.bias Loaded text_branch.encoder.layer.2.attention.output.LayerNorm.weight Loaded text_branch.encoder.layer.2.attention.output.LayerNorm.bias Loaded text_branch.encoder.layer.2.intermediate.dense.weight Loaded text_branch.encoder.layer.2.intermediate.dense.bias Loaded text_branch.encoder.layer.2.output.dense.weight Loaded text_branch.encoder.layer.2.output.dense.bias Loaded text_branch.encoder.layer.2.output.LayerNorm.weight Loaded text_branch.encoder.layer.2.output.LayerNorm.bias Loaded text_branch.encoder.layer.3.attention.self.query.weight Loaded text_branch.encoder.layer.3.attention.self.query.bias Loaded text_branch.encoder.layer.3.attention.self.key.weight Loaded text_branch.encoder.layer.3.attention.self.key.bias Loaded text_branch.encoder.layer.3.attention.self.value.weight Loaded text_branch.encoder.layer.3.attention.self.value.bias Loaded text_branch.encoder.layer.3.attention.output.dense.weight Loaded text_branch.encoder.layer.3.attention.output.dense.bias Loaded text_branch.encoder.layer.3.attention.output.LayerNorm.weight Loaded text_branch.encoder.layer.3.attention.output.LayerNorm.bias Loaded text_branch.encoder.layer.3.intermediate.dense.weight Loaded text_branch.encoder.layer.3.intermediate.dense.bias Loaded text_branch.encoder.layer.3.output.dense.weight Loaded text_branch.encoder.layer.3.output.dense.bias Loaded text_branch.encoder.layer.3.output.LayerNorm.weight Loaded text_branch.encoder.layer.3.output.LayerNorm.bias Loaded text_branch.encoder.layer.4.attention.self.query.weight Loaded text_branch.encoder.layer.4.attention.self.query.bias Loaded text_branch.encoder.layer.4.attention.self.key.weight Loaded text_branch.encoder.layer.4.attention.self.key.bias Loaded text_branch.encoder.layer.4.attention.self.value.weight Loaded text_branch.encoder.layer.4.attention.self.value.bias Loaded text_branch.encoder.layer.4.attention.output.dense.weight Loaded text_branch.encoder.layer.4.attention.output.dense.bias Loaded text_branch.encoder.layer.4.attention.output.LayerNorm.weight Loaded text_branch.encoder.layer.4.attention.output.LayerNorm.bias Loaded text_branch.encoder.layer.4.intermediate.dense.weight Loaded text_branch.encoder.layer.4.intermediate.dense.bias Loaded text_branch.encoder.layer.4.output.dense.weight Loaded text_branch.encoder.layer.4.output.dense.bias Loaded text_branch.encoder.layer.4.output.LayerNorm.weight Loaded text_branch.encoder.layer.4.output.LayerNorm.bias Loaded text_branch.encoder.layer.5.attention.self.query.weight Loaded text_branch.encoder.layer.5.attention.self.query.bias Loaded text_branch.encoder.layer.5.attention.self.key.weight Loaded text_branch.encoder.layer.5.attention.self.key.bias Loaded text_branch.encoder.layer.5.attention.self.value.weight Loaded text_branch.encoder.layer.5.attention.self.value.bias Loaded text_branch.encoder.layer.5.attention.output.dense.weight Loaded text_branch.encoder.layer.5.attention.output.dense.bias Loaded text_branch.encoder.layer.5.attention.output.LayerNorm.weight Loaded text_branch.encoder.layer.5.attention.output.LayerNorm.bias Loaded text_branch.encoder.layer.5.intermediate.dense.weight Loaded text_branch.encoder.layer.5.intermediate.dense.bias Loaded text_branch.encoder.layer.5.output.dense.weight Loaded text_branch.encoder.layer.5.output.dense.bias Loaded text_branch.encoder.layer.5.output.LayerNorm.weight Loaded text_branch.encoder.layer.5.output.LayerNorm.bias Loaded text_branch.encoder.layer.6.attention.self.query.weight Loaded text_branch.encoder.layer.6.attention.self.query.bias Loaded text_branch.encoder.layer.6.attention.self.key.weight Loaded text_branch.encoder.layer.6.attention.self.key.bias Loaded text_branch.encoder.layer.6.attention.self.value.weight Loaded text_branch.encoder.layer.6.attention.self.value.bias Loaded text_branch.encoder.layer.6.attention.output.dense.weight Loaded text_branch.encoder.layer.6.attention.output.dense.bias Loaded text_branch.encoder.layer.6.attention.output.LayerNorm.weight Loaded text_branch.encoder.layer.6.attention.output.LayerNorm.bias Loaded text_branch.encoder.layer.6.intermediate.dense.weight Loaded text_branch.encoder.layer.6.intermediate.dense.bias Loaded text_branch.encoder.layer.6.output.dense.weight Loaded text_branch.encoder.layer.6.output.dense.bias Loaded text_branch.encoder.layer.6.output.LayerNorm.weight Loaded text_branch.encoder.layer.6.output.LayerNorm.bias Loaded text_branch.encoder.layer.7.attention.self.query.weight Loaded text_branch.encoder.layer.7.attention.self.query.bias Loaded text_branch.encoder.layer.7.attention.self.key.weight Loaded text_branch.encoder.layer.7.attention.self.key.bias Loaded text_branch.encoder.layer.7.attention.self.value.weight Loaded text_branch.encoder.layer.7.attention.self.value.bias Loaded text_branch.encoder.layer.7.attention.output.dense.weight Loaded text_branch.encoder.layer.7.attention.output.dense.bias Loaded text_branch.encoder.layer.7.attention.output.LayerNorm.weight Loaded text_branch.encoder.layer.7.attention.output.LayerNorm.bias Loaded text_branch.encoder.layer.7.intermediate.dense.weight Loaded text_branch.encoder.layer.7.intermediate.dense.bias Loaded text_branch.encoder.layer.7.output.dense.weight Loaded text_branch.encoder.layer.7.output.dense.bias Loaded text_branch.encoder.layer.7.output.LayerNorm.weight Loaded text_branch.encoder.layer.7.output.LayerNorm.bias Loaded text_branch.encoder.layer.8.attention.self.query.weight Loaded text_branch.encoder.layer.8.attention.self.query.bias Loaded text_branch.encoder.layer.8.attention.self.key.weight Loaded text_branch.encoder.layer.8.attention.self.key.bias Loaded text_branch.encoder.layer.8.attention.self.value.weight Loaded text_branch.encoder.layer.8.attention.self.value.bias Loaded text_branch.encoder.layer.8.attention.output.dense.weight Loaded text_branch.encoder.layer.8.attention.output.dense.bias Loaded text_branch.encoder.layer.8.attention.output.LayerNorm.weight Loaded text_branch.encoder.layer.8.attention.output.LayerNorm.bias Loaded text_branch.encoder.layer.8.intermediate.dense.weight Loaded text_branch.encoder.layer.8.intermediate.dense.bias Loaded text_branch.encoder.layer.8.output.dense.weight Loaded text_branch.encoder.layer.8.output.dense.bias Loaded text_branch.encoder.layer.8.output.LayerNorm.weight Loaded text_branch.encoder.layer.8.output.LayerNorm.bias Loaded text_branch.encoder.layer.9.attention.self.query.weight Loaded text_branch.encoder.layer.9.attention.self.query.bias Loaded text_branch.encoder.layer.9.attention.self.key.weight Loaded text_branch.encoder.layer.9.attention.self.key.bias Loaded text_branch.encoder.layer.9.attention.self.value.weight Loaded text_branch.encoder.layer.9.attention.self.value.bias Loaded text_branch.encoder.layer.9.attention.output.dense.weight Loaded text_branch.encoder.layer.9.attention.output.dense.bias Loaded text_branch.encoder.layer.9.attention.output.LayerNorm.weight Loaded text_branch.encoder.layer.9.attention.output.LayerNorm.bias Loaded text_branch.encoder.layer.9.intermediate.dense.weight Loaded text_branch.encoder.layer.9.intermediate.dense.bias Loaded text_branch.encoder.layer.9.output.dense.weight Loaded text_branch.encoder.layer.9.output.dense.bias Loaded text_branch.encoder.layer.9.output.LayerNorm.weight Loaded text_branch.encoder.layer.9.output.LayerNorm.bias Loaded text_branch.encoder.layer.10.attention.self.query.weight Loaded text_branch.encoder.layer.10.attention.self.query.bias Loaded text_branch.encoder.layer.10.attention.self.key.weight Loaded text_branch.encoder.layer.10.attention.self.key.bias Loaded text_branch.encoder.layer.10.attention.self.value.weight Loaded text_branch.encoder.layer.10.attention.self.value.bias Loaded text_branch.encoder.layer.10.attention.output.dense.weight Loaded text_branch.encoder.layer.10.attention.output.dense.bias Loaded text_branch.encoder.layer.10.attention.output.LayerNorm.weight Loaded text_branch.encoder.layer.10.attention.output.LayerNorm.bias Loaded text_branch.encoder.layer.10.intermediate.dense.weight Loaded text_branch.encoder.layer.10.intermediate.dense.bias Loaded text_branch.encoder.layer.10.output.dense.weight Loaded text_branch.encoder.layer.10.output.dense.bias Loaded text_branch.encoder.layer.10.output.LayerNorm.weight Loaded text_branch.encoder.layer.10.output.LayerNorm.bias Loaded text_branch.encoder.layer.11.attention.self.query.weight Loaded text_branch.encoder.layer.11.attention.self.query.bias Loaded text_branch.encoder.layer.11.attention.self.key.weight Loaded text_branch.encoder.layer.11.attention.self.key.bias Loaded text_branch.encoder.layer.11.attention.self.value.weight Loaded text_branch.encoder.layer.11.attention.self.value.bias Loaded text_branch.encoder.layer.11.attention.output.dense.weight Loaded text_branch.encoder.layer.11.attention.output.dense.bias Loaded text_branch.encoder.layer.11.attention.output.LayerNorm.weight Loaded text_branch.encoder.layer.11.attention.output.LayerNorm.bias Loaded text_branch.encoder.layer.11.intermediate.dense.weight Loaded text_branch.encoder.layer.11.intermediate.dense.bias Loaded text_branch.encoder.layer.11.output.dense.weight Loaded text_branch.encoder.layer.11.output.dense.bias Loaded text_branch.encoder.layer.11.output.LayerNorm.weight Loaded text_branch.encoder.layer.11.output.LayerNorm.bias Loaded text_branch.pooler.dense.weight Loaded text_branch.pooler.dense.bias Loaded text_transform.sequential.0.weight Loaded text_transform.sequential.0.bias Loaded text_transform.sequential.3.weight Loaded text_transform.sequential.3.bias Loaded text_projection.0.weight Loaded text_projection.0.bias Loaded text_projection.2.weight Loaded text_projection.2.bias Loaded audio_transform.sequential.0.weight Loaded audio_transform.sequential.0.bias Loaded audio_transform.sequential.3.weight Loaded audio_transform.sequential.3.bias Loaded audio_projection.0.weight Loaded audio_projection.0.bias Loaded audio_projection.2.weight Loaded audio_projection.2.bias Loaded [2026-09-06T09:30:40Z] EMBED instrument loaded in 3.0s [2026-09-06T09:30:41Z] EMBED 8/2241 8.23 crops/s eta 4.5 min [2026-09-06T09:31:00Z] EMBED 168/2241 8.30 crops/s eta 4.2 min [2026-09-06T09:31:19Z] EMBED 328/2241 8.32 crops/s eta 3.8 min [2026-09-06T09:31:39Z] EMBED 488/2241 8.29 crops/s eta 3.5 min [2026-09-06T09:31:58Z] EMBED 648/2241 8.28 crops/s eta 3.2 min [2026-09-06T09:32:17Z] EMBED 808/2241 8.28 crops/s eta 2.9 min [2026-09-06T09:32:37Z] EMBED 968/2241 8.28 crops/s eta 2.6 min [2026-09-06T09:32:56Z] EMBED 1128/2241 8.27 crops/s eta 2.2 min [2026-09-06T09:33:16Z] EMBED 1288/2241 8.27 crops/s eta 1.9 min [2026-09-06T09:33:35Z] EMBED 1448/2241 8.24 crops/s eta 1.6 min [2026-09-06T09:33:55Z] EMBED 1608/2241 8.24 crops/s eta 1.3 min [2026-09-06T09:34:14Z] EMBED 1768/2241 8.24 crops/s eta 1.0 min [2026-09-06T09:34:34Z] EMBED 1928/2241 8.24 crops/s eta 0.6 min [2026-09-06T09:34:53Z] EMBED 2088/2241 8.24 crops/s eta 0.3 min [2026-09-06T09:35:12Z] EMBED 2241/2241 8.24 crops/s eta 0.0 min [2026-09-06T09:35:12Z] EMBED 2241 forwards in 272.0s [2026-09-06T09:35:12Z] EMBED crop-vector L2 norms: min=1.0000 max=1.0000 mean=1.0000 [2026-09-06T09:35:12Z] EMBED wrote embeddings.npz 4255369 bytes