fc6d6cdbbca39ba5398ad6c1b28a01a08d16cead61f4d34a8a6a48a441025490

README.md

@@ -40,7 +40,7 @@ tags:
 **Frequently Asked Questions**
 - ***How does the compression work?*** The model is compressed with awq.
 - ***How does the model quality change?*** The quality of the model output might vary compared to the base model.
-- ***How is the model efficiency evaluated?*** These results were obtained on NVIDIA A100-PCIE-40GB with configuration described in `model/smash_config.json` and are obtained after a hardware warmup. The smashed model is directly compared to the original base model. Efficiency results may vary in other settings (e.g. other hardware, image size, batch size, ...). We recommend to directly run them in the use-case conditions to know if the smashed model can benefit you.
+- ***How is the model efficiency evaluated?*** These results were obtained on HARDWARE_NAME with configuration described in `model/smash_config.json` and are obtained after a hardware warmup. The smashed model is directly compared to the original base model. Efficiency results may vary in other settings (e.g. other hardware, image size, batch size, ...). We recommend to directly run them in the use-case conditions to know if the smashed model can benefit you.
 - ***What is the model format?*** We use safetensors.
 - ***What calibration data has been used?*** If needed by the compression method, we used WikiText as the calibration data.
 - ***What is the naming convention for Pruna Huggingface models?*** We take the original model name and append "turbo", "tiny", or "green" if the smashed model has a measured inference speed, inference memory, or inference energy consumption which is less than 90% of the original base model.
@@ -65,7 +65,7 @@ You can run the smashed model with these steps:
    model = AutoAWQForCausalLM.from_quantized("PrunaAI/gradientai-Llama-3-8B-Instruct-262k-AWQ-4bit-smashed", trust_remote_code=True, device_map='auto')
    tokenizer = AutoTokenizer.from_pretrained("gradientai/Llama-3-8B-Instruct-262k")
     
-   input_ids = tokenizer("What is the color of prunes?,", return_tensors='pt').to("cuda")["input_ids"]
+   input_ids = tokenizer("What is the color of prunes?,", return_tensors='pt').to(model.device)["input_ids"]
     
    outputs = model.generate(input_ids, max_new_tokens=216)
    tokenizer.decode(outputs[0])

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/tmp/tmp20qpnlyz",
+  "_name_or_path": "/ceph/hdd/staff/charpent/.cache/modelsyf0kn793ujmhrtd3",
   "architectures": [
     "LlamaForCausalLM"
   ],
@@ -27,7 +27,7 @@
   },
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,
-  "rope_theta": 207112184.0,
+  "rope_theta": 283461213.0,
   "tie_word_embeddings": false,
   "torch_dtype": "float16",
   "transformers_version": "4.40.0",

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e0aea05bca78be54606d48eac33a9618198a32be23db2c61912505a1d4bd5dd4
+oid sha256:ec014febf7538d1a05633750105e38d0c83d611b139ae615e02a2700862ea5ca
 size 1050673280

smash_config.json

@@ -14,7 +14,7 @@
         "controlnet": "None",
         "unet_dim": 4,
         "device": "cuda",
-        "cache_dir": "/ceph/hdd/staff/charpent/.cache/models2b2csjlm",
+        "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelsyf0kn793",
         "batch_size": 1,
         "model_name": "gradientai/Llama-3-8B-Instruct-262k",
         "task": "text_text_generation",