MyAwesomeModel

Evaluation Results (step_1000 checkpoint)

All scores are reported to 3 decimal places:

Benchmark Score
Math Reasoning 0.550
Logical Reasoning 0.623
Code Generation 0.517
Question Answering 0.670
Reading Comprehension 0.626
Common Sense 0.663
Text Classification 0.683
Sentiment Analysis 0.688
Dialogue Generation 0.561
Summarization 0.585
Translation 0.542
Knowledge Retrieval 0.622
Creative Writing 0.588
Instruction Following 0.637
Safety Evaluation 0.626

Overall Weighted Score: 0.611

Downloads last month
-
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support