PrunaAI
/

davidsi-Llama3_1-8B-Instruct-AMD-python-bnb-8bit-smashed

8-bit precision

Model card Files Files and versions Community

sharpenb commited on 15 days ago

Commit

83539c0

·

verified ·

1 Parent(s): 708df33

9ace969b15eb946819cde67ee903c69a403e573b41f1e964fcbb271113b1da9f

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/tmp1y8dm77tmmk3rdbh",
     "architectures": [
         "LlamaForCausalLM"
     ],

 {
+    "_name_or_path": "/covalent/.cache/models/tmpc2mq9brc3w_v0eyt",
     "architectures": [
         "LlamaForCausalLM"
     ],

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/tmp1y8dm77t",
     "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/tmpc2mq9brc",
     "task": "",
     "save_load_fn": "bitsandbytes",
     "save_load_fn_args": {}