Update README.md

README.md

@@ -17,44 +17,36 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [davidkim205/nox-solar-10.7b-v4](https://huggingface.co/davidkim205/nox-solar-10.7b-v4) on an unknown dataset.
 
-## Model description
+### Our Team
+* Youjin Chung
+* Jingyeom Kim
 
-More information needed
+## Model
 
-## Intended uses & limitations
+### Base Model
+* [davidkim205/nox-solar-10.7b-v4](https://huggingface.co/davidkim205/nox-solar-10.7b-v4)
 
-More information needed
+### Hardware and Software
+* Hardware: A100 * 8 for training our model
+* Deepspeed library & Huggingface TRL Trainer
 
-## Training and evaluation data
+### Dataset
+* DPO_dataset
+  * 자체 제작 dpo dataset(AI-hub dataset 활용)
+  * OpenOrca DPO 등 영어 데이터셋 번역(ENERGY-DRINK-LOVE/translate_share_gpt_dedup_llama_SFT_1024, 자체모델 활용)
 
-More information needed
+### Training Method
+* [DPO](https://arxiv.org/abs/2305.18290)
 
-## Training procedure
+## Benchmark
 
-### Training hyperparameters
+**[Ko LM Eval Harness](https://github.com/Beomi/ko-lm-evaluation-harness)**
+### 0 shot (macro f1)
 
-The following hyperparameters were used during training:
-- learning_rate: 5e-07
-- train_batch_size: 1
-- eval_batch_size: 8
-- seed: 42
-- distributed_type: multi-GPU
-- num_devices: 6
-- gradient_accumulation_steps: 8
-- total_train_batch_size: 48
-- total_eval_batch_size: 48
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: cosine
-- lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 1
+| kobest_boolq | kobest_copa | kobest_hellaswag | kobest_sentineg |
+| ------: | -----: | -----------: | ------: |
+|   0.931613 |  0.740751 |        0.468602 |   0.488465 |
 
-### Training results
 
 
 
-### Framework versions
-
-- Transformers 4.36.2
-- Pytorch 2.0.1+cu117
-- Datasets 2.16.1
-- Tokenizers 0.15.1