Apollo🎖️
Collection
Experimental models
•
3 items
•
Updated
Uploaded model
- Developed by: NotAiLOL
- License: apache-2.0
- Finetuned from model : NotAiLOL/Apollo-7b-Experimental
This mistral model was trained 2x faster with Unsloth and Huggingface's TRL library.
Model detail
Prompt format
The model uses the alpaca format:
Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
### Instruction:
{}
### Input:
{}
### Response:
{}
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 71.81 |
| AI2 Reasoning Challenge (25-Shot) | 64.85 |
| HellaSwag (10-Shot) | 85.50 |
| MMLU (5-Shot) | 63.93 |
| TruthfulQA (0-shot) | 63.52 |
| Winogrande (5-shot) | 84.06 |
| GSM8k (5-shot) | 68.99 |
- Downloads last month
- 11
This model does not have enough activity to be deployed to Inference API (serverless) yet.
Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Finetuned from
Dataset used to train NotAiLOL/Apollo-7b-orpo-Experimental
Collection including NotAiLOL/Apollo-7b-orpo-Experimental
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard64.850
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard85.500
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard63.930
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard63.520
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard84.060
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard68.990