{
  "format_version": 1,
  "model_type": "interleaved-lm",
  "architectures": [
    "PerceptionExpressionAdaptedTextLM"
  ],
  "model_args": {
    "txt_vocabsize": 49155,
    "img_vocabsize": null,
    "aud_vocabsize": 502,
    "block_size": 2048,
    "txt_pad_token": 0,
    "img_pad_token": null,
    "aud_pad_token": 501,
    "swt_token": 49154,
    "n_img_codebooks": 0,
    "n_aud_codebooks": 1,
    "backbone": "HuggingFaceTB/SmolLM-1.7B",
    "backbone_revision": "646382037cc9ef78fa50842b0c95fc1cb7a608f9",
    "backbone_config": {
      "vocab_size": 49152,
      "max_position_embeddings": 2048,
      "hidden_size": 2048,
      "intermediate_size": 8192,
      "num_hidden_layers": 24,
      "num_attention_heads": 32,
      "num_key_value_heads": 32,
      "hidden_act": "silu",
      "initializer_range": 0.02,
      "rms_norm_eps": 1e-05,
      "pretraining_tp": 1,
      "use_cache": true,
      "rope_theta": 100000.0,
      "rope_scaling": null,
      "attention_bias": false,
      "attention_dropout": 0.0,
      "mlp_bias": false,
      "head_dim": 64,
      "return_dict": true,
      "output_hidden_states": false,
      "output_attentions": false,
      "torchscript": false,
      "torch_dtype": "float32",
      "use_bfloat16": false,
      "tf_legacy_loss": false,
      "pruned_heads": {},
      "tie_word_embeddings": true,
      "chunk_size_feed_forward": 0,
      "is_encoder_decoder": false,
      "is_decoder": false,
      "cross_attention_hidden_size": null,
      "add_cross_attention": false,
      "tie_encoder_decoder": false,
      "max_length": 20,
      "min_length": 0,
      "do_sample": false,
      "early_stopping": false,
      "num_beams": 1,
      "num_beam_groups": 1,
      "diversity_penalty": 0.0,
      "temperature": 1.0,
      "top_k": 50,
      "top_p": 1.0,
      "typical_p": 1.0,
      "repetition_penalty": 1.0,
      "length_penalty": 1.0,
      "no_repeat_ngram_size": 0,
      "encoder_no_repeat_ngram_size": 0,
      "bad_words_ids": null,
      "num_return_sequences": 1,
      "output_scores": false,
      "return_dict_in_generate": false,
      "forced_bos_token_id": null,
      "forced_eos_token_id": null,
      "remove_invalid_values": false,
      "exponential_decay_length_penalty": null,
      "suppress_tokens": null,
      "begin_suppress_tokens": null,
      "architectures": [
        "LlamaForCausalLM"
      ],
      "finetuning_task": null,
      "id2label": {
        "0": "LABEL_0",
        "1": "LABEL_1"
      },
      "label2id": {
        "LABEL_0": 0,
        "LABEL_1": 1
      },
      "tokenizer_class": null,
      "prefix": null,
      "bos_token_id": 0,
      "pad_token_id": null,
      "eos_token_id": 0,
      "sep_token_id": null,
      "decoder_start_token_id": null,
      "task_specific_params": null,
      "problem_type": null,
      "_name_or_path": "HuggingFaceTB/SmolLM-1.7B",
      "_attn_implementation_autoset": true,
      "transformers_version": "4.48.3",
      "model_type": "llama"
    },
    "tokenizer": "tokenizer",
    "tokenizer_revision": null,
    "warm_init": false,
    "freeze_backbone": false,
    "freeze_txt_inout": false,
    "rope_theta": 100000.0,
    "img_inadapter_n_layers": 6,
    "img_inadapter_dim": 576,
    "img_inadapter_mlp_dim": 1536,
    "img_inadapter_n_heads": 9,
    "img_inadapter_n_kvheads": 3,
    "img_outadapter_n_layers": 6,
    "img_outadapter_dim": 576,
    "img_outadapter_mlp_dim": 1536,
    "img_outadapter_n_heads": 9,
    "img_outadapter_n_kvheads": 3,
    "aud_inadapter_n_layers": 2,
    "aud_inadapter_dim": 2048,
    "aud_inadapter_mlp_dim": 8192,
    "aud_inadapter_n_heads": 32,
    "aud_inadapter_n_kvheads": 32,
    "aud_outadapter_n_layers": 2,
    "aud_outadapter_dim": 2048,
    "aud_outadapter_mlp_dim": 8192,
    "aud_outadapter_n_heads": 32,
    "aud_outadapter_n_kvheads": 32,
    "img_codebook_transformer_layers": 0,
    "img_codebook_transformer_dim": 576,
    "img_codebook_transformer_mlp_dim": 1536,
    "img_codebook_transformer_n_heads": 9,
    "img_codebook_transformer_n_kvheads": 3,
    "aud_codebook_transformer_layers": 0,
    "aud_codebook_transformer_dim": 576,
    "aud_codebook_transformer_mlp_dim": 1536,
    "aud_codebook_transformer_n_heads": 9,
    "aud_codebook_transformer_n_kvheads": 3,
    "img_delay_pattern": null,
    "aud_delay_pattern": null,
    "img_attention_residual": "none",
    "aud_attention_residual": "dynamic",
    "img_attention_residual_entropy_reg": 0.0,
    "aud_attention_residual_entropy_reg": 0.01,
    "img_attention_residual_norm": true,
    "aud_attention_residual_norm": false,
    "img_attention_residual_backbone_only": false,
    "aud_attention_residual_backbone_only": true,
    "tie_img_embeddings": true,
    "tie_aud_embeddings": true,
    "img_head_type": "categorical",
    "aud_head_type": "categorical",
    "img_flow_steps": 20,
    "aud_flow_steps": 20,
    "img_flow_d": 64,
    "aud_flow_d": 64
  },
  "torch_dtype": "float32",
  "weight_aliases": {
    "global_workspace.txt_unembed.weight": "global_workspace.txt_embed.weight",
    "aud_expression.perception_embed.0.weight": "aud_perception.embed.0.weight",
    "aud_expression.unembed.0.weight": "aud_perception.embed.0.weight"
  }
}