config.json

{
  "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,
  "dtype": "bfloat16",
  "eos_token_id": 106,
  "ffn_hidden_size": 8192,
  "fused_add_norm": false,
  "gated_linear_unit": true,
  "hidden_size": 4096,
  "head_dim": 128,
  "kv_channels": 256,
  "lm_head_bias": false,
  "max_position_embeddings": 262144,
  "model_type": "zaya",
  "moe_router_topk": 1,
  "norm_epsilon": 1e-05,
  "normalization": "RMSNorm",
  "num_attention_heads": 16,
  "num_experts": 24,
  "num_hidden_layers": 120,
  "num_key_value_heads": 2,
  "num_query_groups": 2,
  "pad_token_id": 0,
  "residual_in_fp32": true,
  "partial_rotary_factor": 0.5,
  "rope_scaling": false,
  "rope_theta": 10000000,
  "scale_residual_merge": true,
  "sliding_window": null,
  "zaya_high_prec": true,
  "zaya_mlp_expansion": 256,
  "zaya_use_eda": true,
  "zaya_use_mod": true,
  "swa_layers": [
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0,
    4096,
    0,
    0,
    0
  ],
  "swa_rotary_base": 10000,
  "transformers_version": "4.57.1",
  "use_cache": true,
  "vocab_size": 262272
}