{ "architectures": [ "MistralForCausalLM" ], "attention_dropout": 0.0, "bos_token_id": 2, "dtype": "bfloat16", "eos_token_id": 3, "head_dim": 128, "hidden_act": "silu", "hidden_size": 3072, "initializer_range": 0.02, "intermediate_size": 10752, "max_position_embeddings": 8192, "model_name": "Ba2han/experimental3", "model_type": "mistral", "num_attention_heads": 16, "num_hidden_layers": 18, "num_key_value_heads": 4, "pad_token_id": 0, "rms_norm_eps": 1e-05, "rope_parameters": { "rope_theta": 500000, "rope_type": "default" }, "sliding_window": null, "tie_word_embeddings": false, "transformers_version": "5.12.1", "unsloth_version": "2026.6.9", "use_cache": false, "vocab_size": 50176 }