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: 0.4727
-- Rewards/chosen: 0.2480
-- Rewards/rejected: 0.1188
-- Rewards/accuracies: 0.6005
-- Rewards/margins: 0.1292
-- Logps/rejected: -37.3186
-- Logps/chosen: -33.6212
+- Loss: 0.4730
+- Rewards/chosen: 0.2464
+- Rewards/rejected: 0.1175
+- Rewards/accuracies: 0.5947
+- Rewards/margins: 0.1288
+- Logps/rejected: -37.3207
+- Logps/chosen: -33.6239
 - Logits/rejected: -2.1470
 - Logits/chosen: -2.1517
 

all_results.json

@@ -1,17 +1,17 @@
 {
     "epoch": 4.0,
-    "eval_logits/chosen": -2.230694055557251,
-    "eval_logits/rejected": -2.225895643234253,
-    "eval_logps/chosen": -33.75852966308594,
-    "eval_logps/rejected": -37.27712631225586,
-    "eval_loss": 0.4946732223033905,
-    "eval_rewards/accuracies": 0.5365448594093323,
-    "eval_rewards/chosen": 0.16561271250247955,
-    "eval_rewards/margins": 0.02191758342087269,
-    "eval_rewards/rejected": 0.14369513094425201,
-    "eval_runtime": 145.7007,
+    "eval_logits/chosen": -2.1517372131347656,
+    "eval_logits/rejected": -2.1470396518707275,
+    "eval_logps/chosen": -33.62392044067383,
+    "eval_logps/rejected": -37.32072830200195,
+    "eval_loss": 0.47301676869392395,
+    "eval_rewards/accuracies": 0.594684362411499,
+    "eval_rewards/chosen": 0.24637830257415771,
+    "eval_rewards/margins": 0.1288420557975769,
+    "eval_rewards/rejected": 0.11753623187541962,
+    "eval_runtime": 145.7991,
     "eval_samples": 343,
-    "eval_samples_per_second": 2.354,
+    "eval_samples_per_second": 2.353,
     "eval_steps_per_second": 0.295,
     "train_loss": 0.19402528018146367,
     "train_runtime": 10806.8766,

eval_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 1.0,
-    "eval_logits/chosen": -2.230694055557251,
-    "eval_logits/rejected": -2.225895643234253,
-    "eval_logps/chosen": -33.75852966308594,
-    "eval_logps/rejected": -37.27712631225586,
-    "eval_loss": 0.4946732223033905,
-    "eval_rewards/accuracies": 0.5365448594093323,
-    "eval_rewards/chosen": 0.16561271250247955,
-    "eval_rewards/margins": 0.02191758342087269,
-    "eval_rewards/rejected": 0.14369513094425201,
-    "eval_runtime": 145.7007,
+    "epoch": 4.0,
+    "eval_logits/chosen": -2.1517372131347656,
+    "eval_logits/rejected": -2.1470396518707275,
+    "eval_logps/chosen": -33.62392044067383,
+    "eval_logps/rejected": -37.32072830200195,
+    "eval_loss": 0.47301676869392395,
+    "eval_rewards/accuracies": 0.594684362411499,
+    "eval_rewards/chosen": 0.24637830257415771,
+    "eval_rewards/margins": 0.1288420557975769,
+    "eval_rewards/rejected": 0.11753623187541962,
+    "eval_runtime": 145.7991,
     "eval_samples": 343,
-    "eval_samples_per_second": 2.354,
+    "eval_samples_per_second": 2.353,
     "eval_steps_per_second": 0.295
 }