Model Details
- Model Description: This model is test for data ordering.
- Developed by: Juhwan Lee
- Model Type: Large Language Model
Model Architecture
This model is based on Mistral-7B-v0.1. We fine-tuning this model for data ordering task.
Mistral-7B-v0.1 is a transformer model, with the following architecture choices:
- Grouped-Query Attention
- Sliding-Window Attention
- Byte-fallback BPE tokenizer
Dataset
We random sample Open-Orca dataset. (We finetune the 100,000 dataset)
Guthub
License
Apache License 2.0
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 60.80 |
| AI2 Reasoning Challenge (25-Shot) | 60.92 |
| HellaSwag (10-Shot) | 82.92 |
| MMLU (5-Shot) | 61.80 |
| TruthfulQA (0-shot) | 44.69 |
| Winogrande (5-shot) | 78.61 |
| GSM8k (5-shot) | 35.86 |
Dataset used to train juhwanlee/llmdo-Mistral-7B-case-c
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard60.920
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard82.920
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard61.800
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard44.690
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard78.610
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard35.860