Ignis-7B-DPO
Creator: NeuralNovel
Community Organization: ConvexAI
Discord: Join us on Discord
Ignis-7B-DPO Trained on the Neural-DPO dataset using A-100 80GB.
- Downloads last month
- 2,427
Dataset used to train NeuralNovel/Ignis-7B-DPO-Laser
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard65.190
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard84.570
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard58.560
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard66.240
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard80.430
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard31.460