10 epoches

README.md

@@ -37,25 +37,22 @@ Since [gemma-2-2b-jpn-it-ablitered-18](https://huggingface.co/ymcki/gemma-2-2b-j
 
 Using the [gemma-2-2b base model](https://huggingface.co/google/gemma-2-2b), I employed the ORPO method described by [mlabonne](https://towardsdatascience.com/fine-tune-llama-3-with-orpo-56cfab2f9ada) but the input model was read into VRAM by [unsloth](https://github.com/unslothai/unsloth) to allow using the full 40k dataset to run on a single 3090.
 
-Five epoches was run. Smallest eval_loss was achieve at epoch 7.72. 
-Checkpoint at epoch 7.72 is used to obtain a model adapter and
+Five epoches was run. Smallest eval_loss was achieve at epoch 7.00. 
+Checkpoint at epoch 7.00 is used to obtain a model adapter and
 applied it to [gemma-2-2b-jpn-it-ablitered-18](https://huggingface.co/ymcki/gemma-2-2b-jpn-it-abliterated-18) to obtain this model.
 
 | Epoch | loss | eval_loss | eval_logps/rejected | eval_logps/chosen |
 | ----- | ---- | --------- | ------------------- | ----------------- |
-| 1.00 | 1.2868 | 1.0689 | -1.0857 | -0.7500 |
-| 2.00 | 0.9663 | 1.0288 | -1.1321 | -0.7289 |
-| 3.00 | 1.2255 | 1.0297 | -1.1840 | -0.7272 |
-| 4.00 | 1.5293 | 1.0166 | -1.2004 | -0.7200 |
-| 4.96 | 1.2893 | 1.0077 | -1.1754 | -0.7106 |
-| 5.00 | 1.3458 | 1.0078 | -1.1730 | -0.7105 |
-| 6.00 | 1.3807 | 0.9924 | -1.1757 | -0.6971 |
-| 7.00 | 1.0855 | 0.9889 | -1.2634 | -0.7235 |
-| 7.72 | 0.8720 | 0.9855 | -1.2374 | -0.7100 |
-| 8.00 | 0.7301 | 0.9864 | -1.2406 | -0.7113 |
-| 9.00 | 1.1939 | 0.9934 | -1.2703 | -0.6852 |
-| 10.00 | 0.7421 | 1.0269 | -1.2552 | -0.7395 |
-
+| 1.00 | 0.9754 | 1.0344 | -1.1506 | -0.7516 |
+| 2.00 | 0.9629 | 1.0173 | -1.2694 | -0.7351 |
+| 3.00 | 0.7435 | 1.0087 | -1.4922 | -0.7388 |
+| 4.00 | 1.0595 | 1.0026 | -1.5920 | -0.7310 |
+| 5.00 | 1.0525 | 1.0000 | -1.6313 | -0.7311 |
+| 6.00 | 1.1628 | 1.0014 | -1.7263 | -0.7393 |
+| 7.00 | 0.8994 | 0.9971 | -1.7264 | -0.7324 |
+| 8.00 | 0.7448 | 1.0056 | -1.7790 | -0.7482 |
+| 9.00 | 0.6801 | 1.0028 | -1.7794 | -0.7429 |
+| 10.00 | 0.9868 | 1.0069 | -1.8065 | -0.7505 |
 
 This model is uploaded here to be evaluated by the Open LLM Leaderboard. Further ORPO fine tuning is currently underway to see if it can regain its sanity. You can play with this model first or wait until I am done with the fine tuning.
 

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d593d8b3023dd418100c91db48aad44bb3810444dbda5b77c85428538e80fe2
+oid sha256:fb08a3bbd213a780221394190a4c26058b93236368418330307b17fd6e6500d6
 size 4988034976

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b87919ea30834310ead2527bf859d1409e6e52897a4863eae1512617633e91ee
+oid sha256:9939c97a43e37148b7e5a74b7943b79a40961fe859e41474333754003c833e47
 size 240691728