dfurman
/

Llama-3-8B-Orpo-v0.1

Text Generation

text-generation-inference

Inference Endpoints

Model card Files Files and versions Community

dfurman commited on Apr 29

Commit

9203a3b

•

1 Parent(s): 62dbdda

Update README.md

Files changed (1) hide show

README.md +6 -9

README.md CHANGED Viewed

@@ -30,15 +30,12 @@ This model uses a context window of 8k. It was trained with the ChatML template.
 ### Open LLM Leaderboard
-| Metric                | Value                     |
-|-----------------------|---------------------------|
-| Avg.                  | 64.67   |
-| ARC (25-shot)         | 60.67          |
-| HellaSwag (10-shot)   | 82.56    |
-| MMLU (5-shot)         | 66.59         |
-| TruthfulQA (0-shot)   | 50.47   |
-| Winogrande (5-shot)   | 79.01   |
-| GSM8K (5-shot)        | 48.75        |
 ## 📈 Training curves

 ### Open LLM Leaderboard
+| Model                                                                                                                                                                                                       |   Average |   ARC (25-shot) |   HellaSwag (10-shot) | MMLU (5-shot)  |   TruthfulQA (0-shot) |  Winogrande (5-shot) |  GSM8K (5-shot) |
+| ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------: | --------: | --------: | ---------: | --------: |  --------: |  --------: |
+| [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) [📄](https://huggingface.co/datasets/open-llm-leaderboard/details_meta-llama__Meta-Llama-3-8B-Instruct)    |       66.87 |     41.22 |     69.86 |      51.65 |     42.64 |     42.64 |     42.64 |
+| [**dfurman/Llama-3-8B-Orpo-v0.1**](https://huggingface.co/dfurman/Llama-3-8B-Orpo-v0.1) [📄](https://huggingface.co/datasets/open-llm-leaderboard/details_dfurman__Llama-3-8B-Orpo-v0.1)                     | **64.67** | **34.17** | **70.59** | **52.39** | **37.36** |     **42.64** |     **42.64** |
+| [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) [📄](https://huggingface.co/datasets/open-llm-leaderboard/details_meta-llama__Meta-Llama-3-8B)                               |     62.55 |      31.1 |     69.95 |      43.91 |      36.7 |     42.64 |     42.64 |
 ## 📈 Training curves