Kikinoking
/

MNLP_M2_quantized_model

8-bit precision

compressed-tensors

Model card Files Files and versions

Kikinoking commited on May 27

Commit

5c1ea40

·

verified ·

1 Parent(s): 15ee3b9

Upload Qwen3ForCausalLM

Files changed (1) hide show

config.json +2 -2

config.json CHANGED Viewed

@@ -10,7 +10,7 @@
       "input_activations": {
         "num_bits": 8,
         "strategy": "tensor",
-        "type": "float"
       },
       "targets": [
         "Linear"
@@ -18,7 +18,7 @@
       "weights": {
         "num_bits": 8,
         "strategy": "tensor",
-        "type": "float"
       }
     }
   },

       "input_activations": {
         "num_bits": 8,
         "strategy": "tensor",
+        "type": "int"
       },
       "targets": [
         "Linear"
       "weights": {
         "num_bits": 8,
         "strategy": "tensor",
+        "type": "int"
       }
     }
   },