| { | |
| "activation_type": "silu", | |
| "alibi": false, | |
| "alibi_bias_max": 8.0, | |
| "architectures": [ | |
| "LLaDAModelLM" | |
| ], | |
| "attention_dropout": 0.0, | |
| "attention_layer_norm": false, | |
| "attention_layer_norm_with_affine": true, | |
| "auto_map": { | |
| "AutoConfig": "configuration_llada.LLaDAConfig", | |
| "AutoModel": "modeling_llada.LLaDAModelLM", | |
| "AutoModelForCausalLM": "modeling_llada.LLaDAModelLM" | |
| }, | |
| "bias_for_layer_norm": false, | |
| "block_group_size": 1, | |
| "block_type": "llama", | |
| "d_model": 4096, | |
| "embedding_dropout": 0.0, | |
| "embedding_size": 126464, | |
| "eos_token_id": 126081, | |
| "flash_attention": true, | |
| "flash_attn_varlen": true, | |
| "include_bias": false, | |
| "include_qkv_bias": false, | |
| "init_cutoff_factor": null, | |
| "init_device": "meta", | |
| "init_fn": "mitchell", | |
| "init_std": 0.02, | |
| "input_emb_norm": false, | |
| "layer_norm_type": "rms", | |
| "layer_norm_with_affine": true, | |
| "mask_token_id": 126336, | |
| "max_sequence_length": 4096, | |
| "mlp_hidden_size": 12288, | |
| "mlp_ratio": 4, | |
| "model_type": "llada", | |
| "multi_query_attention": null, | |
| "n_heads": 32, | |
| "n_kv_heads": 32, | |
| "n_layers": 32, | |
| "pad_token_id": 126081, | |
| "precision": "amp_bf16", | |
| "residual_dropout": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope": true, | |
| "rope_full_precision": true, | |
| "rope_theta": 500000.0, | |
| "scale_logits": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.51.2", | |
| "use_cache": false, | |
| "vocab_size": 126464, | |
| "weight_tying": false | |
| } | |