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: 0.9713
-- Rewards/chosen: -0.0468
-- Rewards/rejected: -0.0779
-- Rewards/accuracies: 0.5282
-- Rewards/margins: 0.0310
-- Logps/rejected: -37.6032
-- Logps/chosen: -34.0866
-- Logits/rejected: -2.2201
-- Logits/chosen: -2.2249
+- Loss: 0.9479
+- Rewards/chosen: -0.0348
+- Rewards/rejected: -0.0895
+- Rewards/accuracies: 0.5419
+- Rewards/margins: 0.0548
+- Logps/rejected: -37.6161
+- Logps/chosen: -34.0732
+- Logits/rejected: -2.2190
+- Logits/chosen: -2.2239
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -2.223867654800415,
+    "eval_logits/rejected": -2.219041109085083,
+    "eval_logps/chosen": -34.07316970825195,
+    "eval_logps/rejected": -37.6160888671875,
+    "eval_loss": 0.947879433631897,
+    "eval_rewards/accuracies": 0.5419435501098633,
+    "eval_rewards/chosen": -0.0347554087638855,
+    "eval_rewards/margins": 0.05476636067032814,
+    "eval_rewards/rejected": -0.08952176570892334,
+    "eval_runtime": 145.6786,
+    "eval_samples": 343,
+    "eval_samples_per_second": 2.354,
+    "eval_steps_per_second": 0.295,
     "train_loss": 0.8125879040012112,
     "train_runtime": 3249.3137,
     "train_samples": 3079,

eval_results.json

@@ -1,16 +1,16 @@
 {
     "epoch": 1.0,
-    "eval_logits/chosen": -2.238292694091797,
-    "eval_logits/rejected": -2.2334160804748535,
-    "eval_logps/chosen": -34.009239196777344,
-    "eval_logps/rejected": -37.498390197753906,
-    "eval_loss": 0.3797042965888977,
-    "eval_rewards/accuracies": 0.5336378812789917,
-    "eval_rewards/chosen": 0.02277994342148304,
-    "eval_rewards/margins": 0.006376237142831087,
-    "eval_rewards/rejected": 0.01640370488166809,
-    "eval_runtime": 145.6289,
+    "eval_logits/chosen": -2.223867654800415,
+    "eval_logits/rejected": -2.219041109085083,
+    "eval_logps/chosen": -34.07316970825195,
+    "eval_logps/rejected": -37.6160888671875,
+    "eval_loss": 0.947879433631897,
+    "eval_rewards/accuracies": 0.5419435501098633,
+    "eval_rewards/chosen": -0.0347554087638855,
+    "eval_rewards/margins": 0.05476636067032814,
+    "eval_rewards/rejected": -0.08952176570892334,
+    "eval_runtime": 145.6786,
     "eval_samples": 343,
-    "eval_samples_per_second": 2.355,
+    "eval_samples_per_second": 2.354,
     "eval_steps_per_second": 0.295
 }