Starlight
Collection
The Starlight Collection
β’
2 items
β’
Updated
Model | Average β¬οΈ | ARC | HellaSwag | MMLU | TruthfulQA |
---|---|---|---|---|---|
NewstaR/Starlight-13B | 58.63 | 59.3 | 82.15 | 55.67 | 37.39 |
NewstaR/Starlight-7B | 54.3 | 53.07 | 78.57 | 46.8 | 38.75 |
### Instruction: {prompt} ### Response:
### Instruction: Summarize the key details of the Starlight model in a few sentences.
### Response: Starlight is a 13B parameter transformer model trained on the AverageData and Above the Clouds datasets for conversational text generation. It has strong language modeling capabilities but lacks true language understanding and may generate incorrect or biased text, so outputs should be monitored and safeguards implemented. The model is intended for use in chatbots and content creation applications.
Detailed results can be found here
Metric | Value |
---|---|
Avg. | 46.87 |
ARC (25-shot) | 59.3 |
HellaSwag (10-shot) | 82.15 |
MMLU (5-shot) | 55.67 |
TruthfulQA (0-shot) | 37.39 |
Winogrande (5-shot) | 76.64 |
GSM8K (5-shot) | 10.84 |
DROP (3-shot) | 6.08 |