{ "architectures": [ "ARLoopBaselineForCausalLM" ], "auto_map": { "AutoConfig": "configuration_baseline.ARLoopBaselineConfig", "AutoModelForCausalLM": "modeling_baseline.ARLoopBaselineForCausalLM" }, "bos_token_id": 0, "dtype": "bfloat16", "embedding_scale": 1.0, "eos_token_id": 0, "head_dim": 96, "hidden_size": 1536, "intermediate_size": 4608, "logit_softcap": null, "loop_count": 3, "max_position_embeddings": 4096, "model_type": "arloop_baseline", "num_attention_heads": 16, "num_hidden_layers": 54, "num_key_value_heads": 4, "pad_token_id": 0, "physical_layers": 18, "qk_norm": true, "rms_norm_eps": 1e-05, "rope_condense_ratio": 1, "rope_theta": 50000, "tie_word_embeddings": true, "transformers_version": "5.12.1", "use_cache": true, "vocab_size": 49152, "torch_dtype": "bfloat16" }