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: 0.3845
-- Rewards/chosen: 0.0182
-- Rewards/rejected: 0.0177
-- Rewards/accuracies: 0.4934
-- Rewards/margins: 0.0005
-- Logps/rejected: -37.4970
-- Logps/chosen: -34.0143
-- Logits/rejected: -2.2316
-- Logits/chosen: -2.2364
+- Loss: 0.3797
+- Rewards/chosen: 0.0228
+- Rewards/rejected: 0.0164
+- Rewards/accuracies: 0.5336
+- Rewards/margins: 0.0064
+- Logps/rejected: -37.4984
+- Logps/chosen: -34.0092
+- Logits/rejected: -2.2334
+- Logits/chosen: -2.2383
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "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_samples": 343,
+    "eval_samples_per_second": 2.355,
+    "eval_steps_per_second": 0.295,
     "train_loss": 0.31703355428460356,
     "train_runtime": 3251.5033,
     "train_samples": 3079,

eval_results.json

@@ -1,16 +1,16 @@
 {
     "epoch": 1.0,
-    "eval_logits/chosen": -2.2310752868652344,
-    "eval_logits/rejected": -2.226254940032959,
-    "eval_logps/chosen": -33.75144958496094,
-    "eval_logps/rejected": -37.262733459472656,
-    "eval_loss": 0.49370110034942627,
-    "eval_rewards/accuracies": 0.5274086594581604,
-    "eval_rewards/chosen": 0.25479215383529663,
-    "eval_rewards/margins": 0.02629559487104416,
-    "eval_rewards/rejected": 0.22849655151367188,
-    "eval_runtime": 145.6018,
+    "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_samples": 343,
-    "eval_samples_per_second": 2.356,
+    "eval_samples_per_second": 2.355,
     "eval_steps_per_second": 0.295
 }