Create README.md

README.md

@@ -0,0 +1,24 @@
+---
+language: en
+license: apache-2.0
+---
+
+# SQFT: Low-cost Model Adaptation in Low-precision Sparse Foundation Models
+
+- Base Model: [IntelLabs/sqft-phi-3-mini-4k-50-base-gptq](https://huggingface.co/IntelLabs/sqft-phi-3-mini-4k-50-base-gptq)
+- Sparsity: 50%
+- Quantization: INT4 (GPTQ)
+- Finetune Method: SQFT
+- Finetune data: [winogrande](https://huggingface.co/datasets/winogrande), [boolq](https://huggingface.co/datasets/google/boolq), [openbookqa](https://huggingface.co/datasets/allenai/openbookqa), [hellaswag](https://huggingface.co/datasets/Rowan/hellaswag), [piqa](https://huggingface.co/datasets/piqa), [ai2_arc](https://huggingface.co/datasets/allenai/ai2_arc) training dataset (83k)
+- Sub-Adapter: Heuristic
+
+### Evaluation
+
+```bash
+BASE_MODEL_PATH=IntelLabs/sqft-phi-3-mini-4k-50-base-gptq
+ADAPTER_MODEL_PATH=IntelLabs/sqft-phi-3-mini-4k-50-gptq-cs-heu-adapter
+lm_eval --model hf --model_args pretrained=${BASE_MODEL_PATH},peft=${ADAPTER_MODEL_PATH},add_bos_token=True,trust_remote_code=True --tasks piqa,arc_easy,arc_challenge,hellaswag,openbookqa,boolq,winogrande --batch_size auto:4
+```
+
+Refer to our [repo](https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning/tree/main/SQFT) for the environment information to run this command.
+