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: 0.6957
-- Rewards/chosen: -0.0451
-- Rewards/rejected: -0.0645
-- Rewards/accuracies: 0.5399
-- Rewards/margins: 0.0194
-- Logps/rejected: -37.5973
-- Logps/chosen: -34.0909
-- Logits/rejected: -2.2246
-- Logits/chosen: -2.2294
+- Loss: 0.6942
+- Rewards/chosen: -0.0476
+- Rewards/rejected: -0.0693
+- Rewards/accuracies: 0.5100
+- Rewards/margins: 0.0217
+- Logps/rejected: -37.6032
+- Logps/chosen: -34.0940
+- Logits/rejected: -2.2243
+- Logits/chosen: -2.2291
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -2.22914457321167,
+    "eval_logits/rejected": -2.224289894104004,
+    "eval_logps/chosen": -34.094017028808594,
+    "eval_logps/rejected": -37.60322189331055,
+    "eval_loss": 0.6942082643508911,
+    "eval_rewards/accuracies": 0.5099667906761169,
+    "eval_rewards/chosen": -0.047573190182447433,
+    "eval_rewards/margins": 0.021706854924559593,
+    "eval_rewards/rejected": -0.06928005069494247,
+    "eval_runtime": 145.4732,
+    "eval_samples": 343,
+    "eval_samples_per_second": 2.358,
+    "eval_steps_per_second": 0.296,
     "train_loss": 0.6230236078237559,
     "train_runtime": 3254.2307,
     "train_samples": 3079,

eval_results.json

@@ -1,16 +1,16 @@
 {
     "epoch": 1.0,
-    "eval_logits/chosen": -2.2230520248413086,
-    "eval_logits/rejected": -2.2182328701019287,
-    "eval_logps/chosen": -34.07753372192383,
-    "eval_logps/rejected": -37.60981750488281,
-    "eval_loss": 0.9658033847808838,
-    "eval_rewards/accuracies": 0.5278239250183105,
-    "eval_rewards/chosen": -0.03008819743990898,
-    "eval_rewards/margins": 0.03515118733048439,
-    "eval_rewards/rejected": -0.06523937731981277,
-    "eval_runtime": 145.6671,
+    "eval_logits/chosen": -2.22914457321167,
+    "eval_logits/rejected": -2.224289894104004,
+    "eval_logps/chosen": -34.094017028808594,
+    "eval_logps/rejected": -37.60322189331055,
+    "eval_loss": 0.6942082643508911,
+    "eval_rewards/accuracies": 0.5099667906761169,
+    "eval_rewards/chosen": -0.047573190182447433,
+    "eval_rewards/margins": 0.021706854924559593,
+    "eval_rewards/rejected": -0.06928005069494247,
+    "eval_runtime": 145.4732,
     "eval_samples": 343,
-    "eval_samples_per_second": 2.355,
-    "eval_steps_per_second": 0.295
+    "eval_samples_per_second": 2.358,
+    "eval_steps_per_second": 0.296
 }