Upload folder using huggingface_hub (#2)

- f48d6228f47e5dbda3e36a627ff3888ebe08122af2b4499ff771dacc4025d918 (5c67934a11d98a2cf4d591b15944cea3270c5863)

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/cerebras-Cerebras-GPT-590M-bnb-4bit-smashed",
-                                                 trust_remote_code=True)
+                                                 trust_remote_code=True, device_map='auto')
     tokenizer = AutoTokenizer.from_pretrained("cerebras/Cerebras-GPT-590M")
     
     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/tmpo6ey_2o4",
+  "_name_or_path": "/tmp/tmp2kj4wb0y",
   "activation_function": "gelu",
   "architectures": [
     "GPT2LMHeadModel"
@@ -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 sha256:61d5f20119417da516c68b7043f0aa59097ed64e68705229b9f885167b70b12e
-size 424450684
+oid sha256:a615d4f81d3394e0caa4f686eababc7ee4985255c52c42511df48740592a71c0
+size 448118504

smash_config.json

@@ -8,7 +8,7 @@
         "compilers": "None",
         "task": "text_text_generation",
         "device": "cuda",
-        "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelsbpl6ndbz",
+        "cache_dir": "/ceph/hdd/staff/charpent/.cache/models59fr1xzu",
         "batch_size": 1,
         "model_name": "cerebras/Cerebras-GPT-590M",
         "pruning_ratio": 0.0,