Text Generation
MLX
mistral
awni's picture
Add rope theta for quantized model
8d170fd
raw
history blame contribute delete
304 Bytes
{
"dim": 4096,
"n_layers": 32,
"head_dim": 128,
"hidden_dim": 14336,
"n_heads": 32,
"n_kv_heads": 8,
"norm_eps": 1e-05,
"vocab_size": 32000,
"rope_theta": 1000000.0,
"quantization": {
"group_size": 64,
"bits": 4
},
"model_type": "mistral"
}