Upload folder using huggingface_hub (#2)

- a5b606556584bb1e3dfb7eee7028b0c03f123e0586c1899a90d99930c9d5e990 (2c585179e151c6a603cebc27a421ff410433168d)
- b561ecb4877ef9b83e0dc24f44137a341331d2df7ba33eecaaed030dea4f017d (1c8a466f79e1cf7e1345bc4052c1348011fc17e3)

README.md

@@ -34,7 +34,7 @@ tags:
 
 ## Results
 
-Detailed efficiency metrics coming soon!
+![image info](./plots.png)
 
 **Frequently Asked Questions**
 - ***How does the compression work?*** The model is compressed with llm-int8.
@@ -61,7 +61,7 @@ You can run the smashed model with these steps:
    from transformers import AutoModelForCausalLM, AutoTokenizer
 
     model = AutoModelForCausalLM.from_pretrained("PrunaAI/norallm-normistral-7b-warm-bnb-4bit-smashed",
-                                                 trust_remote_code=True)
+                                                 trust_remote_code=True, device_map='auto')
     tokenizer = AutoTokenizer.from_pretrained("norallm/normistral-7b-warm")
     
     input_ids = tokenizer("What is the color of prunes?,", return_tensors='pt').to(model.device)["input_ids"]

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/tmp/tmpxjmp01z9",
+  "_name_or_path": "/tmp/tmpqh5ssxox",
   "architectures": [
     "MistralForCausalLM"
   ],
@@ -19,7 +19,7 @@
   "quantization_config": {
     "bnb_4bit_compute_dtype": "bfloat16",
     "bnb_4bit_quant_type": "fp4",
-    "bnb_4bit_use_double_quant": true,
+    "bnb_4bit_use_double_quant": false,
     "llm_int8_enable_fp32_cpu_offload": false,
     "llm_int8_has_fp16_weight": false,
     "llm_int8_skip_modules": [

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a75963210e2e50ee6088e87276919528dba8b7dde4a102ed5eedc8cc6685e1a
-size 4138270407
+oid sha256:c0dab75c3caa755abe42ab657cc18177389b9f459030ac81ccbb1e4d9c53dda2
+size 4463420416

smash_config.json

@@ -8,7 +8,7 @@
         "compilers": "None",
         "task": "text_text_generation",
         "device": "cuda",
-        "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelsxs8bxeqn",
+        "cache_dir": "/ceph/hdd/staff/charpent/.cache/models_0l9t9wr",
         "batch_size": 1,
         "model_name": "norallm/normistral-7b-warm",
         "pruning_ratio": 0.0,