{
  "architectures": [
    "NeuroBLASTForCausalLM"
  ],
  "auto_map": {
    "AutoConfig": "configuration_neuroblast.NeuroBLASTConfig",
    "AutoModelForCausalLM": "modeling_neuroblast.NeuroBLASTForCausalLM"
  },
  "attention_bias": false,
  "attention_dropout": 0.0,
  "attention_every": 0,
  "dropout": 0.0,
  "dtype": "float32",
  "hidden_act": "silu",
  "hidden_size": 512,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "kernel_size": 5,
  "layer_types": [
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention",
    "full_attention"
  ],
  "max_position_embeddings": 32768,
  "model_type": "neuroblast",
  "num_associative_layers": 32,
  "num_attention_heads": 16,
  "num_hidden_layers": 72,
  "num_key_value_heads": 8,
  "num_motor_layers": 16,
  "num_sensory_layers": 24,
  "pad_token_id": 65537,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "scale": 1.0,
  "sliding_window": null,
  "temporal_kernel_size": 5,
  "tie_word_embeddings": false,
  "transformers_version": "4.57.1",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 65538
}