mlx-community
/

Qwen2-72B-4bit

Text Generation

Model card Files Files and versions Community

prince-canuma commited on 20 days ago

Commit

a85ed31

•

1 Parent(s): cc0e1dd

Update config.json

Files changed (1) hide show

config.json +1 -1

config.json CHANGED Viewed

@@ -17,7 +17,7 @@
     "num_key_value_heads": 8,
     "quantization": {
         "group_size": 64,
-        "bits": 8
     },
     "rms_norm_eps": 1e-05,
     "rope_theta": 1000000.0,

     "num_key_value_heads": 8,
     "quantization": {
         "group_size": 64,
+        "bits": 4
     },
     "rms_norm_eps": 1e-05,
     "rope_theta": 1000000.0,