metadata
base_model: appvoid/arco
language:
- en
license: apache-2.0
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
- sft
Prompt
Similar to the popular llama3-70b-reflection model you can prompt it as follows:
What is 12 + 12?
<thinking>
| Task | Score | Metric |
|---|---|---|
| ARC Challenge | 0.3541 | acc_norm |
| HellaSwag | 0.6049 | acc_norm |
| MMLU | 0.2730 | acc |
| PIQA | 0.7247 | acc_norm |
| Winogrande | 0.6022 | acc |
This table presents the extracted scores in a clear, tabular format. The "Task" column shows the name of each benchmark, the "Score" column displays the corresponding value, and the "Metric" column indicates whether the score is acc_norm or acc.
Uploaded model
- Developed by: appvoid
- License: apache-2.0
- Finetuned from model : appvoid/arco
This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.
