{ "architectures": [ "CacaForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "attn_logit_softcapping": null, "auto_map": { "AutoConfig": "configuration_caca.CacaConfig", "AutoModel": "modeling_caca.CacaModel", "AutoModelForCausalLM": "modeling_caca.CacaForCausalLM" }, "bos_token_id": 2, "dtype": "bfloat16", "embedding_dropout": 0.0, "eos_token_id": 1, "final_logit_softcapping": null, "gradient_checkpointing": false, "head_dim": 128, "hidden_activation": "silu", "hidden_dropout": 0.0, "hidden_size": 1280, "initializer_range": 0.02, "intermediate_size": 5120, "kv_lora_rank": 213, "layer_types": [ "sliding_attention", "full_attention", "sliding_attention", "full_attention", "sliding_attention", "full_attention", "sliding_attention", "full_attention", "sliding_attention", "full_attention", "sliding_attention", "full_attention", "sliding_attention", "full_attention", "sliding_attention", "full_attention", "sliding_attention", "full_attention", "sliding_attention", "full_attention" ], "max_position_embeddings": 8192, "model_type": "caca", "mtp_loss_weight": 0.3, "num_attention_heads": 10, "num_hidden_layers": 20, "num_key_value_heads": 8, "num_mtp_tokens": 1, "pad_token_id": 0, "q_lora_rank": 426, "qk_nope_head_dim": 64, "qk_rope_head_dim": 32, "query_pre_attn_scalar": 128, "rms_norm_eps": 1e-06, "rope_parameters": { "beta_fast": 32, "beta_slow": 1, "factor": 4.0, "original_max_position_embeddings": 8192, "rope_theta": 10000.0, "rope_type": "yarn" }, "sliding_window": 4096, "tie_word_embeddings": true, "transformers_version": "5.13.1", "use_cache": true, "use_mla": true, "use_qk_norm": true, "v_head_dim": 128, "vocab_size": 32000 }