PrunaAI
/

tiiuae-Falcon3-7B-Instruct-bnb-8bit-smashed

8-bit precision

Model card Files Files and versions Community

Upload folder using huggingface_hub

#2

by sharpenb - opened 9 days ago

base: refs/heads/main

←

from: refs/pr/2

Discussion Files changed

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/tmpx5o910l6krbovpd3",
     "architectures": [
         "LlamaForCausalLM"
     ],

 {
+    "_name_or_path": "/covalent/.cache/models/tmpimn2nijtizynm7v2",
     "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/tmpx5o910l6",
     "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/tmpimn2nijt",
     "task": "",
     "save_load_fn": "bitsandbytes",
     "save_load_fn_args": {}