Upload folder using huggingface_hub (#4)

- 295c5a668e9ce17e4c7469391fa827492768f9468ebf9a7ccbcbfeeceb92be24 (67c215fdbc2bef46fae17492c62fbd784fac5be3)

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/akreal-tiny-random-BloomForCausalLM-bnb-4bit-smashed",
-                                                 trust_remote_code=True)
+                                                 trust_remote_code=True, device_map='auto')
     tokenizer = AutoTokenizer.from_pretrained("akreal/tiny-random-BloomForCausalLM")
     
     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/tmphezyhe0h",
+  "_name_or_path": "/tmp/tmpg7ax3ftc",
   "apply_residual_connection_post_layernorm": false,
   "architectures": [
     "BloomForCausalLM"
@@ -33,7 +33,7 @@
   "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 sha256:6049ac992e3a0b80d7a824043921c74a7efd92b398a498bde6d6e7f041d13dca
-size 145680
+oid sha256:13dbd4c6366040994c6af447bb94eb258b417dc93bc5664672947b51e6a78d6a
+size 121190

smash_config.json

@@ -8,7 +8,7 @@
         "compilers": "None",
         "task": "text_text_generation",
         "device": "cuda",
-        "cache_dir": "/ceph/hdd/staff/charpent/.cache/models1_vonx1x",
+        "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelsjxkv6xfl",
         "batch_size": 1,
         "model_name": "akreal/tiny-random-BloomForCausalLM",
         "pruning_ratio": 0.0,