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: 0.9828
-- Rewards/chosen: -0.0112
-- Rewards/rejected: -0.0284
-- Rewards/accuracies: 0.5395
-- Rewards/margins: 0.0172
-- Logps/rejected: -37.5734
-- Logps/chosen: -34.0569
-- Logits/rejected: -2.2180
-- Logits/chosen: -2.2228
+- Loss: 0.9846
+- Rewards/chosen: -0.0137
+- Rewards/rejected: -0.0292
+- Rewards/accuracies: 0.5137
+- Rewards/margins: 0.0155
+- Logps/rejected: -37.5751
+- Logps/chosen: -34.0619
+- Logits/rejected: -2.2177
+- Logits/chosen: -2.2225
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -2.2225441932678223,
+    "eval_logits/rejected": -2.217723846435547,
+    "eval_logps/chosen": -34.06193161010742,
+    "eval_logps/rejected": -37.57505416870117,
+    "eval_loss": 0.9846408367156982,
+    "eval_rewards/accuracies": 0.5137043595314026,
+    "eval_rewards/chosen": -0.013691178523004055,
+    "eval_rewards/margins": 0.015524487011134624,
+    "eval_rewards/rejected": -0.02921566367149353,
+    "eval_runtime": 145.6349,
+    "eval_samples": 343,
+    "eval_samples_per_second": 2.355,
+    "eval_steps_per_second": 0.295,
     "train_loss": 0.8806782920639237,
     "train_runtime": 3251.4322,
     "train_samples": 3079,

eval_results.json

@@ -1,16 +1,16 @@
 {
     "epoch": 1.0,
-    "eval_logits/chosen": -2.232933521270752,
-    "eval_logits/rejected": -2.2280824184417725,
-    "eval_logps/chosen": -34.02503967285156,
-    "eval_logps/rejected": -37.51841735839844,
-    "eval_loss": 1.0816813707351685,
-    "eval_rewards/accuracies": 0.5103820562362671,
-    "eval_rewards/chosen": 0.004756772890686989,
-    "eval_rewards/margins": 0.005652438849210739,
-    "eval_rewards/rejected": -0.0008956658421084285,
-    "eval_runtime": 145.5936,
+    "eval_logits/chosen": -2.2225441932678223,
+    "eval_logits/rejected": -2.217723846435547,
+    "eval_logps/chosen": -34.06193161010742,
+    "eval_logps/rejected": -37.57505416870117,
+    "eval_loss": 0.9846408367156982,
+    "eval_rewards/accuracies": 0.5137043595314026,
+    "eval_rewards/chosen": -0.013691178523004055,
+    "eval_rewards/margins": 0.015524487011134624,
+    "eval_rewards/rejected": -0.02921566367149353,
+    "eval_runtime": 145.6349,
     "eval_samples": 343,
-    "eval_samples_per_second": 2.356,
+    "eval_samples_per_second": 2.355,
     "eval_steps_per_second": 0.295
 }