TODO: check the splits of each dataset
Loss curves:
- Downloads last month
- 4
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
Model tree for BrownianNotion/TinyLlama_v1.1_mix_wikitext_alpaca_2bit_BitDistiller_baseline
Base model
TinyLlama/TinyLlama_v1.1Datasets used to train BrownianNotion/TinyLlama_v1.1_mix_wikitext_alpaca_2bit_BitDistiller_baseline
Evaluation results
- Accuracy on ARC-Challengetest set self-reported0.215
- Normalized Accuracy on ARC-Challengetest set self-reported0.246
- Accuracy on HellaSwagtest set self-reported0.324
- Normalized Accuracy on HellaSwagtest set self-reported0.373
- Accuracy on PIQAvalidation set self-reported0.608
- Normalized Accuracy on PIQAvalidation set self-reported0.607
- Accuracy on Winograndetest set self-reported0.520
- QA Average on QA-Avgself-reported0.417
- Perplexity on WikiText-2test set self-reported22.655