{ "architectures": [ "LLM" ], "context_length": 1024, "dtype": "float32", "emb_dim": 768, "hidden_dim": 3072, "is_moe": false, "model_type": "custom_llm", "n_experts": null, "n_heads": 12, "n_layers": 12, "qkv_bias": false, "top_k": null, "transformers_version": "5.15.0", "vocab_size": 50257 }