End of training

README.md

@@ -1,11 +1,13 @@
 ---
 library_name: peft
 tags:
+- alignment-handbook
 - trl
 - dpo
-- alignment-handbook
 - generated_from_trainer
 base_model: NbAiLab/nb-gpt-j-6B-v2
+datasets:
+- hugodk-sch/aftonposten_title_prefs
 model-index:
 - name: aftonposten-6b-align-scan
   results: []
@@ -16,15 +18,15 @@ should probably proofread and complete it, then remove this comment. -->
 
 # aftonposten-6b-align-scan
 
-This model is a fine-tuned version of [NbAiLab/nb-gpt-j-6B-v2](https://huggingface.co/NbAiLab/nb-gpt-j-6B-v2) on an unknown dataset.
+This model is a fine-tuned version of [data/ap-gpt-j-6b-sft-qlora-04-08](https://huggingface.co/data/ap-gpt-j-6b-sft-qlora-04-08) on the hugodk-sch/aftonposten_title_prefs dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7352
-- Rewards/chosen: 0.0537
-- Rewards/rejected: 0.0336
-- Rewards/accuracies: 0.5399
-- Rewards/margins: 0.0201
-- Logps/rejected: -37.4606
-- Logps/chosen: -33.9451
+- Loss: 0.7456
+- Rewards/chosen: 0.0514
+- Rewards/rejected: 0.0332
+- Rewards/accuracies: 0.5365
+- Rewards/margins: 0.0182
+- Logps/rejected: -37.4614
+- Logps/chosen: -33.9489
 - Logits/rejected: -2.2407
 - Logits/chosen: -2.2456
 

all_results.json

@@ -1,17 +1,17 @@
 {
     "epoch": 4.0,
-    "eval_logits/chosen": -2.2331926822662354,
-    "eval_logits/rejected": -2.2283315658569336,
-    "eval_logps/chosen": -34.061973571777344,
-    "eval_logps/rejected": -37.566593170166016,
-    "eval_loss": 0.7597638368606567,
-    "eval_rewards/accuracies": 0.5307309031486511,
-    "eval_rewards/chosen": -0.016451681032776833,
-    "eval_rewards/margins": 0.013532036915421486,
-    "eval_rewards/rejected": -0.029983719810843468,
-    "eval_runtime": 145.5395,
+    "eval_logits/chosen": -2.2455787658691406,
+    "eval_logits/rejected": -2.2406980991363525,
+    "eval_logps/chosen": -33.94891357421875,
+    "eval_logps/rejected": -37.46136474609375,
+    "eval_loss": 0.7456009984016418,
+    "eval_rewards/accuracies": 0.5365448594093323,
+    "eval_rewards/chosen": 0.051383573561906815,
+    "eval_rewards/margins": 0.018230266869068146,
+    "eval_rewards/rejected": 0.03315330296754837,
+    "eval_runtime": 145.63,
     "eval_samples": 343,
-    "eval_samples_per_second": 2.357,
+    "eval_samples_per_second": 2.355,
     "eval_steps_per_second": 0.295,
     "train_loss": 0.2286405209984098,
     "train_runtime": 10803.5829,

eval_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 1.0,
-    "eval_logits/chosen": -2.2331926822662354,
-    "eval_logits/rejected": -2.2283315658569336,
-    "eval_logps/chosen": -34.061973571777344,
-    "eval_logps/rejected": -37.566593170166016,
-    "eval_loss": 0.7597638368606567,
-    "eval_rewards/accuracies": 0.5307309031486511,
-    "eval_rewards/chosen": -0.016451681032776833,
-    "eval_rewards/margins": 0.013532036915421486,
-    "eval_rewards/rejected": -0.029983719810843468,
-    "eval_runtime": 145.5395,
+    "epoch": 4.0,
+    "eval_logits/chosen": -2.2455787658691406,
+    "eval_logits/rejected": -2.2406980991363525,
+    "eval_logps/chosen": -33.94891357421875,
+    "eval_logps/rejected": -37.46136474609375,
+    "eval_loss": 0.7456009984016418,
+    "eval_rewards/accuracies": 0.5365448594093323,
+    "eval_rewards/chosen": 0.051383573561906815,
+    "eval_rewards/margins": 0.018230266869068146,
+    "eval_rewards/rejected": 0.03315330296754837,
+    "eval_runtime": 145.63,
     "eval_samples": 343,
-    "eval_samples_per_second": 2.357,
+    "eval_samples_per_second": 2.355,
     "eval_steps_per_second": 0.295
 }