PrunaAI
/

qingy2024-GRMR-2B-Instruct-bnb-8bit-smashed

8-bit precision

Model card Files Files and versions

sharpenb commited on Dec 11, 2024

Commit

f4cc932

·

verified ·

1 Parent(s): 7a626ed

50fcb6acff161fa45e73b2610f50ea785882b6e02478da31b3fe58677cf902c6

Files changed (2) hide show

config.json +1 -1
smash_config.json +1 -1

config.json CHANGED Viewed

@@ -1,5 +1,5 @@
 {
-    "_name_or_path": "/covalent/.cache/models/tmp_r44o6bhsd7hewz2",
     "architectures": [
         "Gemma2ForCausalLM"
     ],

 {
+    "_name_or_path": "/covalent/.cache/models/tmp_uvx646a_7wmftn4",
     "architectures": [
         "Gemma2ForCausalLM"
     ],

smash_config.json CHANGED Viewed

@@ -28,7 +28,7 @@
     "quant_llm-int8_weight_bits": 8,
     "max_batch_size": 1,
     "device": "cuda",
-    "cache_dir": "/covalent/.cache/models/tmp_r44o6bh",
     "task": "",
     "save_load_fn": "bitsandbytes",
     "save_load_fn_args": {}

     "quant_llm-int8_weight_bits": 8,
     "max_batch_size": 1,
     "device": "cuda",
+    "cache_dir": "/covalent/.cache/models/tmp_uvx646a",
     "task": "",
     "save_load_fn": "bitsandbytes",
     "save_load_fn_args": {}