End of training

README.md

@@ -1,10 +1,13 @@
 ---
 library_name: peft
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
 base_model: NorLLM-AI/NorMistral-7B
+datasets:
+- hugodk-sch/aftonposten_title_prefs
 model-index:
 - name: norllm-ai-normistral-7b-align-scan
   results: []
@@ -15,17 +18,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # norllm-ai-normistral-7b-align-scan
 
-This model is a fine-tuned version of [NorLLM-AI/NorMistral-7B](https://huggingface.co/NorLLM-AI/NorMistral-7B) on the None dataset.
+This model is a fine-tuned version of [data/norllm-ai-normistral-7b-sft-qlora](https://huggingface.co/data/norllm-ai-normistral-7b-sft-qlora) on the hugodk-sch/aftonposten_title_prefs dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6302
-- Rewards/chosen: 0.0524
-- Rewards/rejected: 0.0270
-- Rewards/accuracies: 0.5311
-- Rewards/margins: 0.0254
-- Logps/rejected: -34.6596
-- Logps/chosen: -31.2076
-- Logits/rejected: -2.8071
-- Logits/chosen: -2.8097
+- Loss: 0.6292
+- Rewards/chosen: 0.0553
+- Rewards/rejected: 0.0226
+- Rewards/accuracies: 0.5166
+- Rewards/margins: 0.0327
+- Logps/rejected: -34.6659
+- Logps/chosen: -31.2035
+- Logits/rejected: -2.8063
+- Logits/chosen: -2.8088
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -2.808791399002075,
+    "eval_logits/rejected": -2.8062777519226074,
+    "eval_logps/chosen": -31.203502655029297,
+    "eval_logps/rejected": -34.66593933105469,
+    "eval_loss": 0.629190981388092,
+    "eval_rewards/accuracies": 0.5166113376617432,
+    "eval_rewards/chosen": 0.05526338145136833,
+    "eval_rewards/margins": 0.03271109238266945,
+    "eval_rewards/rejected": 0.022552290931344032,
+    "eval_runtime": 112.7274,
+    "eval_samples": 343,
+    "eval_samples_per_second": 3.043,
+    "eval_steps_per_second": 0.381,
     "train_loss": 1.2163990609057538,
     "train_runtime": 2720.9807,
     "train_samples": 3079,

eval_results.json

@@ -1,16 +1,16 @@
 {
     "epoch": 1.0,
-    "eval_logits/chosen": -2.821664571762085,
-    "eval_logits/rejected": -2.819110155105591,
-    "eval_logps/chosen": -31.076623916625977,
-    "eval_logps/rejected": -34.71748352050781,
-    "eval_loss": 0.49079883098602295,
-    "eval_rewards/accuracies": 0.5867940187454224,
-    "eval_rewards/chosen": 0.04116562753915787,
-    "eval_rewards/margins": 0.045030828565359116,
-    "eval_rewards/rejected": -0.003865197068080306,
-    "eval_runtime": 112.8844,
+    "eval_logits/chosen": -2.808791399002075,
+    "eval_logits/rejected": -2.8062777519226074,
+    "eval_logps/chosen": -31.203502655029297,
+    "eval_logps/rejected": -34.66593933105469,
+    "eval_loss": 0.629190981388092,
+    "eval_rewards/accuracies": 0.5166113376617432,
+    "eval_rewards/chosen": 0.05526338145136833,
+    "eval_rewards/margins": 0.03271109238266945,
+    "eval_rewards/rejected": 0.022552290931344032,
+    "eval_runtime": 112.7274,
     "eval_samples": 343,
-    "eval_samples_per_second": 3.039,
+    "eval_samples_per_second": 3.043,
     "eval_steps_per_second": 0.381
 }