End of training

README.md

@@ -18,13 +18,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1345
-- Eval/rewards/chosen: 4.0853
-- Eval/logps/chosen: -162.6833
-- Eval/rewards/rejected: 4.3450
-- Eval/logps/rejected: -171.7619
-- Eval/rewards/margins: -0.2597
-- Eval/kl: 37.7215
+- Loss: 0.9779
+- Eval/rewards/chosen: -0.1420
+- Eval/logps/chosen: -204.9554
+- Eval/rewards/rejected: -0.2358
+- Eval/logps/rejected: -217.5698
+- Eval/rewards/margins: 0.0939
+- Eval/kl: 0.0017
 
 ## Model description
 
@@ -43,7 +43,7 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0001
+- learning_rate: 1e-05
 - train_batch_size: 1
 - eval_batch_size: 2
 - seed: 5678
@@ -56,14 +56,14 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss |         |
-|:-------------:|:-----:|:----:|:---------------:|:-------:|
-| 0.6644        | 0.96  | 12   | 0.8564          | 0.0143  |
-| 0.7117        | 2.0   | 25   | 1.0975          | 20.1969 |
-| 0.3864        | 2.96  | 37   | 1.1568          | 33.5785 |
-| 0.4414        | 4.0   | 50   | 1.1371          | 34.2847 |
-| 0.347         | 4.96  | 62   | 1.1333          | 37.5375 |
-| 0.3002        | 5.76  | 72   | 1.1345          | 37.7215 |
+| Training Loss | Epoch | Step | Validation Loss |        |
+|:-------------:|:-----:|:----:|:---------------:|:------:|
+| 0.7513        | 0.96  | 12   | 1.0249          | 0.0985 |
+| 0.9525        | 2.0   | 25   | 1.0135          | 0.0195 |
+| 0.7253        | 2.96  | 37   | 0.9964          | 0.0117 |
+| 0.8924        | 4.0   | 50   | 0.9858          | 0.0097 |
+| 0.7014        | 4.96  | 62   | 0.9809          | 0.0136 |
+| 0.6514        | 5.76  | 72   | 0.9779          | 0.0017 |
 
 
 ### Framework versions

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:25fb1e85f3783f19cfdf18c575d3196fb7e5ff9247059a148562a454bb36a9f5
+oid sha256:4fc67192acce172526cb049fef2d05ddf1304efe649c5d02ec47eaca5686b251
 size 27297544

metrics.jsonl

@@ -76,3 +76,9 @@
 {"epoch": 4.0, "precision": 0.6999999965, "recall": 0.7777777734567901, "fold": 0}
 {"epoch": 4.96, "precision": 0.5999999976, "recall": 0.8333333287037037, "fold": 0}
 {"epoch": 5.76, "precision": 0.6799999972799999, "recall": 0.9444444391975308, "fold": 0}
+{"epoch": 0.96, "precision": 0.34042553119058394, "recall": 0.8888888839506173, "fold": 0}
+{"epoch": 2.0, "precision": 0.37499999921875, "recall": 0.9999999944444444, "fold": 0}
+{"epoch": 2.96, "precision": 0.35999999928, "recall": 0.9999999944444444, "fold": 0}
+{"epoch": 4.0, "precision": 0.3333333326797386, "recall": 0.9444444391975308, "fold": 0}
+{"epoch": 4.96, "precision": 0.3478260862003781, "recall": 0.8888888839506173, "fold": 0}
+{"epoch": 5.76, "precision": 0.3673469380258226, "recall": 0.9999999944444444, "fold": 0}

metrics_epoch_0.96_fold_0_lr_1e-05_seed_5678_weight_5.0.json

@@ -0,0 +1 @@
+{"epoch": 0.96, "precision": 0.34042553119058394, "recall": 0.8888888839506173, "fold": 0}

metrics_epoch_2.0_fold_0_lr_1e-05_seed_5678_weight_5.0.json

@@ -0,0 +1 @@
+{"epoch": 2.0, "precision": 0.37499999921875, "recall": 0.9999999944444444, "fold": 0}

metrics_epoch_2.96_fold_0_lr_1e-05_seed_5678_weight_5.0.json

@@ -0,0 +1 @@
+{"epoch": 2.96, "precision": 0.35999999928, "recall": 0.9999999944444444, "fold": 0}

metrics_epoch_4.0_fold_0_lr_1e-05_seed_5678_weight_5.0.json

@@ -0,0 +1 @@
+{"epoch": 4.0, "precision": 0.3333333326797386, "recall": 0.9444444391975308, "fold": 0}

metrics_epoch_4.96_fold_0_lr_1e-05_seed_5678_weight_5.0.json

@@ -0,0 +1 @@
+{"epoch": 4.96, "precision": 0.3478260862003781, "recall": 0.8888888839506173, "fold": 0}

metrics_epoch_5.76_fold_0_lr_1e-05_seed_5678_weight_5.0.json

@@ -0,0 +1 @@
+{"epoch": 5.76, "precision": 0.3673469380258226, "recall": 0.9999999944444444, "fold": 0}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6947aa77e83d5bc53e9c13d61abccbd998cdcb09a4da2bc29364f223286e7a23
+oid sha256:0a3e5c02b9e2d6b945fd3552f9850aa91e7f136b0a542c17a843e83e6c54c20d
 size 5688