Adding Evaluation Results
#4
by leaderboard-pr-bot - opened
README.md
CHANGED
|
@@ -7,4 +7,17 @@ language:
|
|
| 7 |
pipeline_tag: conversational
|
| 8 |
---
|
| 9 |
# LLaMA 2 Wizard 70B QLoRA
|
| 10 |
-
Fine tuned on WizardLM/WizardLM_evol_instruct_V2_196k dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
pipeline_tag: conversational
|
| 8 |
---
|
| 9 |
# LLaMA 2 Wizard 70B QLoRA
|
| 10 |
+
Fine tuned on WizardLM/WizardLM_evol_instruct_V2_196k dataset.
|
| 11 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 12 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_v2ray__LLaMA-2-Wizard-70B-QLoRA)
|
| 13 |
+
|
| 14 |
+
| Metric | Value |
|
| 15 |
+
|-----------------------|---------------------------|
|
| 16 |
+
| Avg. | 65.4 |
|
| 17 |
+
| ARC (25-shot) | 67.58 |
|
| 18 |
+
| HellaSwag (10-shot) | 87.52 |
|
| 19 |
+
| MMLU (5-shot) | 69.11 |
|
| 20 |
+
| TruthfulQA (0-shot) | 61.79 |
|
| 21 |
+
| Winogrande (5-shot) | 82.32 |
|
| 22 |
+
| GSM8K (5-shot) | 30.48 |
|
| 23 |
+
| DROP (3-shot) | 59.03 |
|