MyAwesomeModel
This is the best checkpoint selected from the workspace based on eval_accuracy.
- Selected checkpoint:
step_1000 - eval_accuracy:
0.704
Evaluation
eval_accuracy is the weighted average benchmark score computed from the available workspace evaluation logic.
| Benchmark | Score |
|---|---|
| math_reasoning | 0.550 |
| code_generation | 0.650 |
| text_classification | 0.828 |
| sentiment_analysis | 0.792 |
| question_answering | 0.607 |
| logical_reasoning | 0.622 |
| common_sense | 0.758 |
| reading_comprehension | 0.811 |
| dialogue_generation | 0.667 |
| summarization | 0.655 |
| translation | 0.775 |
| knowledge_retrieval | 0.625 |
| creative_writing | 0.694 |
| instruction_following | 0.844 |
| safety_evaluation | 0.753 |
- Downloads last month
- 32