Xenova's picture
Xenova HF Staff
Upload folder using huggingface_hub
0189238 verified
{
"activation_func": "swiglu",
"activation_func_fp8_input_store": false,
"add_bias_linear": false,
"architectures": [
"ZayaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bias_activation_fusion": true,
"bos_token_id": 2,
"cca": true,
"cca_time0": 2,
"cca_time1": 2,
"clamp_temp": false,
"dtype": "bfloat16",
"eos_token_id": 106,
"ffn_hidden_size": 64,
"gated_linear_unit": true,
"head_dim": 32,
"hidden_size": 32,
"kv_channels": 128,
"lm_head_bias": false,
"mamba_cache_dtype": "float32",
"max_position_embeddings": 2048,
"model_type": "zaya",
"moe_router_topk": 1,
"norm_epsilon": 1e-05,
"normalization": "RMSNorm",
"num_attention_heads": 2,
"num_experts": 2,
"num_hidden_layers": 4,
"num_key_value_heads": 1,
"num_query_groups": 1,
"pad_token_id": 0,
"partial_rotary_factor": 0.5,
"residual_in_fp32": true,
"rope_parameters": {
"factor": 1.0,
"partial_rotary_factor": 0.5,
"rope_theta": 5000000,
"rope_type": "linear"
},
"rope_theta": 5000000,
"scale_residual_merge": true,
"sliding_window": null,
"swa_layers": null,
"swa_rotary_base": null,
"tie_word_embeddings": true,
"transformers_version": "5.8.0.dev0",
"use_cache": true,
"vocab_size": 262272,
"zaya_high_prec": true,
"zaya_mlp_expansion": 16,
"zaya_use_eda": true,
"zaya_use_mod": true,
"transformers.js_config": {
"use_external_data_format": true
}
}