8e260d3c9b68167d931ac3fa33c35ebb87f5ce6dcb258b5aa55859bd18898973

README.md

@@ -1,6 +1,6 @@
 ---
 thumbnail: "https://assets-global.website-files.com/646b351987a8d8ce158d1940/64ec9e96b4334c0e1ac41504_Logo%20with%20white%20text.svg"
-base_model: McGill-NLP/Llama-3-8B-Web
+base_model: ORIGINAL_REPO_NAME
 metrics:
 - memory_disk
 - memory_inference
@@ -31,7 +31,7 @@ tags:
 - Contact us and tell us which model to compress next [here](https://www.pruna.ai/contact).
 - Request access to easily compress your *own* AI models [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).
 - Read the documentations to know more [here](https://pruna-ai-pruna.readthedocs-hosted.com/en/latest/)
-- Join Pruna AI community on Discord [here](https://discord.gg/rskEr4BZJx) to share feedback/suggestions or get help.
+- Join Pruna AI community on Discord [here](https://discord.gg/CP4VSgck) to share feedback/suggestions or get help.
 
 ## Results
 
@@ -40,7 +40,7 @@ tags:
 **Frequently Asked Questions**
 - ***How does the compression work?*** The model is compressed with llm-int8.
 - ***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 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.
@@ -52,7 +52,7 @@ tags:
 
 You can run the smashed model with these steps:
 
-0. Check requirements from the original repo McGill-NLP/Llama-3-8B-Web installed. In particular, check python, cuda, and transformers versions.
+0. Check requirements from the original repo ORIGINAL_REPO_NAME installed. In particular, check python, cuda, and transformers versions.
 1. Make sure that you have installed quantization related packages.
     ```bash
     pip install transformers accelerate bitsandbytes>0.37.0
@@ -63,7 +63,7 @@ You can run the smashed model with these steps:
    
 
    model = AutoModelForCausalLM.from_pretrained("PrunaAI/McGill-NLP-Llama-3-8B-Web-bnb-4bit-smashed", trust_remote_code=True, device_map='auto')
-   tokenizer = AutoTokenizer.from_pretrained("McGill-NLP/Llama-3-8B-Web")
+   tokenizer = AutoTokenizer.from_pretrained("ORIGINAL_REPO_NAME")
     
    input_ids = tokenizer("What is the color of prunes?,", return_tensors='pt').to(model.device)["input_ids"]
     
@@ -77,7 +77,7 @@ The configuration info are in `smash_config.json`.
 
 ## Credits & License
 
-The license of the smashed model follows the license of the original model. Please check the license of the original model McGill-NLP/Llama-3-8B-Web before using this model which provided the base model. The license  of the `pruna-engine` is [here](https://pypi.org/project/pruna-engine/) on Pypi.
+The license of the smashed model follows the license of the original model. Please check the license of the original model ORIGINAL_REPO_NAME before using this model which provided the base model. The license  of the `pruna-engine` is [here](https://pypi.org/project/pruna-engine/) on Pypi.
 
 ## Want to compress other models?
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/tmp/tmpzkd7tnsk",
+  "_name_or_path": "/tmp/models/tmpjyjfbg6q06azzt4z",
   "architectures": [
     "LlamaForCausalLM"
   ],
@@ -7,11 +7,13 @@
   "attention_dropout": 0.0,
   "bos_token_id": 128000,
   "eos_token_id": 128001,
+  "head_dim": 128,
   "hidden_act": "silu",
   "hidden_size": 4096,
   "initializer_range": 0.02,
   "intermediate_size": 14336,
   "max_position_embeddings": 8192,
+  "mlp_bias": false,
   "model_type": "llama",
   "num_attention_heads": 32,
   "num_hidden_layers": 32,
@@ -38,8 +40,8 @@
   "rope_scaling": null,
   "rope_theta": 500000.0,
   "tie_word_embeddings": false,
-  "torch_dtype": "float16",
-  "transformers_version": "4.40.0",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.48.2",
   "use_cache": true,
   "vocab_size": 128256
 }

generation_config.json

@@ -5,5 +5,5 @@
     128001,
     128009
   ],
-  "transformers_version": "4.40.0"
+  "transformers_version": "4.48.2"
 }

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc7f7c99a906f9b3130e120dd1e28803e9337de4aca7bd53a043247264cb32bd
+oid sha256:ec8e7d6d47280e40ba4e12c5f32ef4c0c83fb322e864679233deabc324d0a1da
 size 1050673280

model.safetensors.index.json

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_size": 6027779904
+    "total_size": 6027780128
   },
   "weight_map": {
     "lm_head.weight": "model-00002-of-00002.safetensors",

smash_config.json

@@ -1,31 +1,23 @@
 {
-    "api_key": null,
-    "verify_url": "http://johnrachwan.pythonanywhere.com",
-    "smash_config": {
-        "pruners": "None",
-        "pruning_ratio": 0.0,
-        "factorizers": "None",
-        "quantizers": "['llm-int8']",
-        "weight_quantization_bits": 4,
-        "output_deviation": 0.005,
-        "compilers": "None",
-        "static_batch": true,
-        "static_shape": true,
-        "controlnet": "None",
-        "unet_dim": 4,
-        "device": "cuda",
-        "cache_dir": "/ceph/hdd/staff/charpent/.cache/models4gd1n_a8",
-        "batch_size": 1,
-        "model_name": "McGill-NLP/Llama-3-8B-Web",
-        "task": "text_text_generation",
-        "max_batch_size": 1,
-        "qtype_weight": "torch.qint8",
-        "qtype_activation": "torch.quint8",
-        "qobserver": "<class 'torch.ao.quantization.observer.MinMaxObserver'>",
-        "qscheme": "torch.per_tensor_symmetric",
-        "qconfig": "x86",
-        "group_size": 128,
-        "damp_percent": 0.1,
-        "save_load_fn": "bitsandbytes"
-    }
+    "batchers": null,
+    "cachers": null,
+    "compilers": null,
+    "distillers": null,
+    "pruners": null,
+    "quantizers": "llm-int8",
+    "recoverers": null,
+    "quant_llm-int8_compute_dtype": "bfloat16",
+    "quant_llm-int8_double_quant": false,
+    "quant_llm-int8_enable_fp32_cpu_offload": false,
+    "quant_llm-int8_has_fp16_weight": false,
+    "quant_llm-int8_quant_type": "fp4",
+    "quant_llm-int8_threshold": 6.0,
+    "quant_llm-int8_weight_bits": 4,
+    "max_batch_size": 1,
+    "device": "cuda",
+    "cache_dir": "/tmp/models/tmpjyjfbg6q",
+    "task": "",
+    "save_load_fn": "llm-int8",
+    "save_load_fn_args": {},
+    "api_key": null
 }