config.json

{
  "activation_function": "gelu_new",
  "architectures": [
    "TinyRecursiveModel"
  ],
  "attn_pdrop": 0.1,
  "dtype": "float32",
  "embd_pdrop": 0.1,
  "hidden_size": 640,
  "layer_norm_epsilon": 1e-05,
  "max_position_embeddings": 1024,
  "model_type": "recursive_gpt",
  "n_embd": 640,
  "n_head": 8,
  "n_inner": null,
  "n_loops": 8,
  "n_physical_layers": 3,
  "n_positions": 1024,
  "num_attention_heads": 8,
  "num_hidden_layers": 3,
  "reorder_and_upcast_attn": false,
  "resid_pdrop": 0.1,
  "scale_attn_by_inverse_layer_idx": false,
  "scale_attn_weights": true,
  "transformers_version": "4.56.0",
  "vocab_size": 50257
}