{ "architectures": [ "Nae1ForCausalLM" ], "attn_pdrop": 0.1, "bos_token_id": 2, "dtype": "float32", "embd_pdrop": 0.1, "eos_token_id": 3, "head_dim": 64, "hidden_act": "silu", "intermediate_size": 2048, "layer_norm_epsilon": 1e-05, "model_type": "nae1", "n_embd": 768, "n_head": 12, "n_kv_heads": 4, "n_layer": 12, "n_positions": 2048, "n_query_groups": 3, "pad_token_id": 0, "resid_pdrop": 0.1, "rope_theta": 1000000.0, "tie_word_embeddings": true, "transformers_version": "5.16.1", "vocab_size": 32000 }