PrunaAI
/

Amden-Llama-2-7b-chat-finetune-resume-a-d-bnb-8bit-smashed

8-bit precision

Model card Files Files and versions Community

sharpenb commited on 18 days ago

Commit

b4f93e3

·

verified ·

1 Parent(s): 8e9f390

ca01d0534b98898c3fc4a29b3779648b928ac37090e25df95f6e7f3f64873338

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/tmpdx7wossevxec31zh",
     "architectures": [
         "LlamaForCausalLM"
     ],

 {
+    "_name_or_path": "/covalent/.cache/models/tmpule6fztvok0qjyex",
     "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/tmpdx7wosse",
     "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/tmpule6fztv",
     "task": "",
     "save_load_fn": "bitsandbytes",
     "save_load_fn_args": {}