Upload folder using huggingface_hub (#3)

- b2773d857c2b8d2f4a0f47482d7c4fa12acad754c10f1d81c119d95592ee7c9e (d5d5b9190a4889ab5ddf91155cad0eb88c5705d3)
- ddc3950c5079bdfcc0982ea768f05d19c4756b87f3327c1d55d5fdd06a21e050 (b8ff36fe279e6be726f2f67619f6a95b349be5b8)

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/Nondzu-zephyr-speakleash-007-pl-8192-32-16-0.05-bnb-4bit-smashed",
-                                                 trust_remote_code=True)
+                                                 trust_remote_code=True, device_map='auto')
     tokenizer = AutoTokenizer.from_pretrained("Nondzu/zephyr-speakleash-007-pl-8192-32-16-0.05")
     
     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/tmpvg386on4",
+  "_name_or_path": "/tmp/tmpk7gvk_li",
   "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 sha256:16a09472438661f249f2f832e2e3e383a2ca3678faa57ffdfe1fadbb9b0a5a01
-size 4125687617
+oid sha256:7b7e57e4c5493e3944d7f74ed100bb28c0ea054f80e8fb8613c541b2388089f8
+size 4450837496

smash_config.json

@@ -8,7 +8,7 @@
         "compilers": "None",
         "task": "text_text_generation",
         "device": "cuda",
-        "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelsm61jlut0",
+        "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelsfzmolb9g",
         "batch_size": 1,
         "model_name": "Nondzu/zephyr-speakleash-007-pl-8192-32-16-0.05",
         "pruning_ratio": 0.0,