Tim-05369/orpo_train_using_space_demo

README.md

@@ -17,18 +17,18 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-Chat-v1.0](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.8073
-- Rewards/chosen: -0.0719
-- Rewards/rejected: -0.1594
+- Loss: 2.2772
+- Rewards/chosen: -0.1134
+- Rewards/rejected: -0.0902
 - Rewards/accuracies: 0.6667
-- Rewards/margins: 0.0874
-- Logps/rejected: -3.1874
-- Logps/chosen: -1.4385
-- Logits/rejected: -2.2859
-- Logits/chosen: -2.4167
-- Nll Loss: 1.7887
-- Log Odds Ratio: -0.3720
-- Log Odds Chosen: 2.6873
+- Rewards/margins: -0.0232
+- Logps/rejected: -1.8039
+- Logps/chosen: -2.2683
+- Logits/rejected: -2.3984
+- Logits/chosen: -2.3109
+- Nll Loss: 2.2241
+- Log Odds Ratio: -1.0632
+- Log Odds Chosen: -0.4972
 
 ## Model description
 
@@ -48,8 +48,8 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 8
-- eval_batch_size: 8
+- train_batch_size: 32
+- eval_batch_size: 32
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:74ddbbdceecb03ec6e89c6ae9a82ff27449b7190a035a4c92a99d41dc6a682c0
+oid sha256:e342c4848f6e6352a74f56ba62320f875186805671cf32b49f37cd80bca8d59c
 size 4400216536

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ccba72281dc9a290f89ffb1ade8e85412a27e156a73f32f98f07081e5a0b7c57
+oid sha256:e59fa5bdfb1ef5d310d364528550c13174df64b6cdcd5d935757d10e0bc07d39
 size 5496