| { |
| "activation_func": "swiglu", |
| "activation_func_fp8_input_store": false, |
| "add_bias_linear": false, |
| "apply_rope_fusion": true, |
| "architectures": [ |
| "ZayaForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "bias_activation_fusion": true, |
| "bos_token_id": 2, |
| "cca": true, |
| "cca_num_q_heads": [ |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0, |
| 8, |
| 0 |
| ], |
| "eos_token_id": 1, |
| "ffn_hidden_size_list": [ |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096, |
| 0, |
| 4096 |
| ], |
| "fused_add_norm": false, |
| "gated_linear_unit": true, |
| "hidden_size": 2048, |
| "kv_channels": 128, |
| "lm_head_bias": false, |
| "max_position_embeddings": 32768, |
| "model_type": "zaya", |
| "moe_router_topk": 1, |
| "norm_epsilon": 1e-05, |
| "normalization": "RMSNorm", |
| "num_attention_heads": 16, |
| "num_hidden_layers": 120, |
| "num_key_value_heads": 2, |
| "num_query_groups_list": [ |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0, |
| 2, |
| 0 |
| ], |
| "pad_token_id": 0, |
| "residual_in_fp32": false, |
| "partial_rotary_factor": 0.5, |
| "rope_scaling": false, |
| "rope_theta": 1000000, |
| "scale_residual_merge": true, |
| "sliding_window": null, |
| "zaya_high_prec": true, |
| "zaya_layers": [ |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16, |
| "a", |
| 16 |
| ], |
| "zaya_mlp_expansion": [ |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256, |
| 0, |
| 256 |
| ], |
| "zaya_use_eda": true, |
| "zaya_use_mod": true, |
| "torch_dtype": "bfloat16", |
| "transformers_version": "4.57.1", |
| "use_cache": true, |
| "vocab_size": 262272 |
| } |
|
|