End of training

README.md

@@ -1,10 +1,13 @@
 ---
 library_name: peft
 tags:
+- alignment-handbook
 - trl
 - dpo
 - 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: []
@@ -15,17 +18,17 @@ 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: 0.5010
-- Rewards/chosen: -0.0014
-- Rewards/rejected: 0.0035
-- Rewards/accuracies: 0.4431
-- Rewards/margins: -0.0049
-- Logps/rejected: -37.4991
-- Logps/chosen: -34.0418
-- Logits/rejected: -2.2397
-- Logits/chosen: -2.2446
+- Loss: 0.5001
+- Rewards/chosen: 0.0029
+- Rewards/rejected: 0.0032
+- Rewards/accuracies: 0.4842
+- Rewards/margins: -0.0003
+- Logps/rejected: -37.5006
+- Logps/chosen: -34.0202
+- Logits/rejected: -2.2393
+- Logits/chosen: -2.2442
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -2.244190216064453,
+    "eval_logits/rejected": -2.2393178939819336,
+    "eval_logps/chosen": -34.020179748535156,
+    "eval_logps/rejected": -37.50062561035156,
+    "eval_loss": 0.500054657459259,
+    "eval_rewards/accuracies": 0.4842192530632019,
+    "eval_rewards/chosen": 0.002875381149351597,
+    "eval_rewards/margins": -0.0003236339834984392,
+    "eval_rewards/rejected": 0.0031990152783691883,
+    "eval_runtime": 145.749,
+    "eval_samples": 343,
+    "eval_samples_per_second": 2.353,
+    "eval_steps_per_second": 0.295,
     "train_loss": 0.4997496530607149,
     "train_runtime": 3257.6696,
     "train_samples": 3079,

eval_results.json

@@ -1,15 +1,15 @@
 {
     "epoch": 1.0,
-    "eval_logits/chosen": -2.244539499282837,
-    "eval_logits/rejected": -2.239666700363159,
-    "eval_logps/chosen": -33.99818801879883,
-    "eval_logps/rejected": -37.488731384277344,
-    "eval_loss": 0.4998374581336975,
-    "eval_rewards/accuracies": 0.5191029906272888,
-    "eval_rewards/chosen": 0.0036363049875944853,
-    "eval_rewards/margins": 0.0008474953356198967,
-    "eval_rewards/rejected": 0.0027888098265975714,
-    "eval_runtime": 145.763,
+    "eval_logits/chosen": -2.244190216064453,
+    "eval_logits/rejected": -2.2393178939819336,
+    "eval_logps/chosen": -34.020179748535156,
+    "eval_logps/rejected": -37.50062561035156,
+    "eval_loss": 0.500054657459259,
+    "eval_rewards/accuracies": 0.4842192530632019,
+    "eval_rewards/chosen": 0.002875381149351597,
+    "eval_rewards/margins": -0.0003236339834984392,
+    "eval_rewards/rejected": 0.0031990152783691883,
+    "eval_runtime": 145.749,
     "eval_samples": 343,
     "eval_samples_per_second": 2.353,
     "eval_steps_per_second": 0.295