ainz/tiny-recursive-model
545
1{2 "activation_function": "gelu_new",3 "architectures": [4 "TinyRecursiveModel"5 ],6 "attn_pdrop": 0.1,7 "dtype": "float32",8 "embd_pdrop": 0.1,9 "hidden_size": 640,10 "layer_norm_epsilon": 1e-05,11 "max_position_embeddings": 1024,12 "model_type": "recursive_gpt",13 "n_embd": 640,14 "n_head": 8,15 "n_inner": null,16 "n_loops": 8,17 "n_physical_layers": 3,18 "n_positions": 1024,19 "num_attention_heads": 8,20 "num_hidden_layers": 3,21 "reorder_and_upcast_attn": false,22 "resid_pdrop": 0.1,23 "scale_attn_by_inverse_layer_idx": false,24 "scale_attn_weights": true,25 "transformers_version": "4.56.0",26 "vocab_size": 5025727}28 