Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,16 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: llama2
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: llama2
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
The LLaMA-2-7b model finetuned on the Math task using [CorDA](arxiv.org/pdf/2406.05223) in the KPA mode with nqopen.
|
| 6 |
+
|
| 7 |
+
| Method | TriviaQA | NQ open | GSM8k | Math |
|
| 8 |
+
|---|---|---|---|---|
|
| 9 |
+
|LoRA|44.17|1.91|42.68|5.92|
|
| 10 |
+
|[CorDA (KPA with nqopen)](https://huggingface.co/iboing/CorDA_KPA_nqopen_finetuned_math/tree/main) | **45.23** | **10.44** | 45.64 | 6.94|
|
| 11 |
+
|[CorDA (IPA with MetaMath)](https://huggingface.co/iboing/CorDA_IPA_math_finetuned_math/tree/main) | - | - | **54.59** | **8.54** |
|
| 12 |
+
|
| 13 |
+
You can evaluate the model's performance following the step-3 in [CorDA github repo](https://github.com/iboing/CorDA).
|
| 14 |
+
|
| 15 |
+
Note: The model trained using CorDA adapter is based on customized code. If you want to restore the original LLaMA architecture, execute `merge_adapter_for_corda.py` in [CorDA github repo](https://github.com/iboing/CorDA).
|
| 16 |
+
|