mamba-130m / config.json
Wauplin's picture
Wauplin HF staff
Push model using huggingface_hub.
72ca876 verified
raw
history blame
274 Bytes
{
"attn_cfg": {},
"attn_layer_idx": [],
"d_intermediate": 0,
"d_model": 768,
"fused_add_norm": true,
"n_layer": 24,
"pad_vocab_size_multiple": 8,
"residual_in_fp32": true,
"rms_norm": true,
"ssm_cfg": {},
"tie_embeddings": true,
"vocab_size": 50277
}