Upload folder using huggingface_hub (#2)

- 1154b82314499f96e482a9c0cc83cff3a9c4b043be5f030942862b1717a094b2 (cb967d8e651a6216bb28763c553b84fa02894719)
- 7f1aaca061d85d4407ba0b32486ca905111ebb170e1bcea4f21eca4aeee11171 (0ce4e5f4960795c47eda0edec3e584ef63fdf37d)

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/BioMistral-BioMistral-7B-bnb-4bit-smashed",
-                                                 trust_remote_code=True)
+                                                 trust_remote_code=True, device_map='auto')
     tokenizer = AutoTokenizer.from_pretrained("BioMistral/BioMistral-7B")
     
     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/tmpkytv6ws8",
+  "_name_or_path": "/tmp/tmpm9nzsnah",
   "architectures": [
     "MistralForCausalLM"
   ],
@@ -18,7 +18,7 @@
   "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 sha256:545ce9485cf5e9d8f784c62136e76728bcd9a7493465beada51a578b9dcb564d
-size 4125687617
+oid sha256:3e8cee80f52325b4e1cafbf2e095347c292d01b85efedf86d1c27bf41f231a77
+size 4450837496

smash_config.json

@@ -8,7 +8,7 @@
         "compilers": "None",
         "task": "text_text_generation",
         "device": "cuda",
-        "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelsagmp5113",
+        "cache_dir": "/ceph/hdd/staff/charpent/.cache/models6cu4vkhr",
         "batch_size": 1,
         "model_name": "BioMistral/BioMistral-7B",
         "pruning_ratio": 0.0,