Gemma 2B Zephyr SFT
The Zephyr SFT recipe applied on top of Gemma 2B
Model description
- Model type: A 2.5B parameter GPT-like model fine-tuned on a mix of publicly available, synthetic datasets.
- Language(s) (NLP): Primarily English
- Finetuned from model: google/gemma-7b
Recipe
We trained using the alignment handbook recipe and logging to W&B
Visit the W&B workspace here
License
This model has the same license as the original Gemma model collection
Compute provided by Lambda Labs - 8xA100 80GB node
- Around 2 hours to train
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
Metric | Value |
---|---|
Avg. | 47.18 |
AI2 Reasoning Challenge (25-Shot) | 49.74 |
HellaSwag (10-Shot) | 72.38 |
MMLU (5-Shot) | 41.37 |
TruthfulQA (0-shot) | 34.42 |
Winogrande (5-shot) | 66.93 |
GSM8k (5-shot) | 18.27 |
- Downloads last month
- 1,809
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for wandb/gemma-2b-zephyr-sft
Base model
google/gemma-2b
Finetuned
this model
Dataset used to train wandb/gemma-2b-zephyr-sft
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard49.740
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard72.380
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard41.370
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard34.420
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard66.930
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard18.270