| { | |
| "_name_or_path": "configs/gated_deltaproduct_1B_l17.json", | |
| "allow_neg_eigval": false, | |
| "architectures": [ | |
| "GatedDeltaProductForCausalLM" | |
| ], | |
| "attn": null, | |
| "attn_mode": "chunk", | |
| "bos_token_id": 1, | |
| "conv_size": 4, | |
| "eos_token_id": 2, | |
| "expand_k": 1.0, | |
| "expand_v": 1.0, | |
| "fuse_cross_entropy": true, | |
| "fuse_norm": true, | |
| "fuse_swiglu": true, | |
| "head_dim": 256, | |
| "hidden_act": "swish", | |
| "hidden_ratio": 4, | |
| "hidden_size": 2048, | |
| "initializer_range": 0.02, | |
| "intermediate_size": null, | |
| "max_position_embeddings": 2048, | |
| "model_type": "gated_deltaproduct", | |
| "norm_eps": 1e-06, | |
| "num_heads": 8, | |
| "num_hidden_layers": 17, | |
| "num_householder": 4, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.48.2", | |
| "use_cache": true, | |
| "use_forget_gate": true, | |
| "use_gate": false, | |
| "use_short_conv": true, | |
| "vocab_size": 32000 | |
| } | |