Adding Evaluation Results
#108
by
leaderboard-pr-bot
- opened
README.md
CHANGED
@@ -230,4 +230,17 @@ To learn more about the pretraining dataset, see the 📓 [RefinedWeb paper](htt
|
|
230 |
Falcon-7B-Instruct is made available under the Apache 2.0 license.
|
231 |
|
232 |
## Contact
|
233 |
-
falconllm@tii.ae
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
230 |
Falcon-7B-Instruct is made available under the Apache 2.0 license.
|
231 |
|
232 |
## Contact
|
233 |
+
falconllm@tii.ae
|
234 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
235 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_tiiuae__falcon-7b-instruct)
|
236 |
+
|
237 |
+
| Metric | Value |
|
238 |
+
|-----------------------|---------------------------|
|
239 |
+
| Avg. | 37.47 |
|
240 |
+
| ARC (25-shot) | 46.16 |
|
241 |
+
| HellaSwag (10-shot) | 70.85 |
|
242 |
+
| MMLU (5-shot) | 25.84 |
|
243 |
+
| TruthfulQA (0-shot) | 44.08 |
|
244 |
+
| Winogrande (5-shot) | 67.96 |
|
245 |
+
| GSM8K (5-shot) | 1.9 |
|
246 |
+
| DROP (3-shot) | 5.54 |
|