End of training

README.md

@@ -1,11 +1,13 @@
 ---
-license: apache-2.0
 library_name: peft
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
 base_model: norallm/normistral-7b-warm
+datasets:
+- hugodk-sch/aftonposten_title_prefs
 model-index:
 - name: ap-normistral-7b-align-scan
   results: []
@@ -16,17 +18,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # ap-normistral-7b-align-scan
 
-This model is a fine-tuned version of [norallm/normistral-7b-warm](https://huggingface.co/norallm/normistral-7b-warm) on the None dataset.
+This model is a fine-tuned version of [data/ap-normistral-7b-sft-qlora](https://huggingface.co/data/ap-normistral-7b-sft-qlora) on the hugodk-sch/aftonposten_title_prefs dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0124
-- Rewards/chosen: -0.0636
-- Rewards/rejected: -0.1409
-- Rewards/accuracies: 0.5399
-- Rewards/margins: 0.0773
-- Logps/rejected: -36.1679
-- Logps/chosen: -32.5340
-- Logits/rejected: 98.7098
-- Logits/chosen: 98.7339
+- Loss: 1.0237
+- Rewards/chosen: -0.0663
+- Rewards/rejected: -0.1206
+- Rewards/accuracies: 0.5311
+- Rewards/margins: 0.0543
+- Logps/rejected: -36.1388
+- Logps/chosen: -32.5378
+- Logits/rejected: 98.6966
+- Logits/chosen: 98.7185
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": 98.71852111816406,
+    "eval_logits/rejected": 98.69664001464844,
+    "eval_logps/chosen": -32.537837982177734,
+    "eval_logps/rejected": -36.13877868652344,
+    "eval_loss": 1.023661732673645,
+    "eval_rewards/accuracies": 0.531146228313446,
+    "eval_rewards/chosen": -0.06626088172197342,
+    "eval_rewards/margins": 0.054315000772476196,
+    "eval_rewards/rejected": -0.12057589739561081,
+    "eval_runtime": 103.7609,
+    "eval_samples": 343,
+    "eval_samples_per_second": 3.306,
+    "eval_steps_per_second": 0.414,
     "train_loss": 0.7429099231571347,
     "train_runtime": 2555.3187,
     "train_samples": 3079,

eval_results.json

@@ -1,16 +1,16 @@
 {
     "epoch": 1.0,
-    "eval_logits/chosen": 98.98882293701172,
-    "eval_logits/rejected": 98.9751968383789,
-    "eval_logps/chosen": -32.437740325927734,
-    "eval_logps/rejected": -35.96149826049805,
-    "eval_loss": 8.017751693725586,
-    "eval_rewards/accuracies": 0.5220099687576294,
-    "eval_rewards/chosen": 0.0010875342413783073,
-    "eval_rewards/margins": 8.220935706049204e-05,
-    "eval_rewards/rejected": 0.0010053252335637808,
-    "eval_runtime": 103.7392,
+    "eval_logits/chosen": 98.71852111816406,
+    "eval_logits/rejected": 98.69664001464844,
+    "eval_logps/chosen": -32.537837982177734,
+    "eval_logps/rejected": -36.13877868652344,
+    "eval_loss": 1.023661732673645,
+    "eval_rewards/accuracies": 0.531146228313446,
+    "eval_rewards/chosen": -0.06626088172197342,
+    "eval_rewards/margins": 0.054315000772476196,
+    "eval_rewards/rejected": -0.12057589739561081,
+    "eval_runtime": 103.7609,
     "eval_samples": 343,
     "eval_samples_per_second": 3.306,
-    "eval_steps_per_second": 0.415
+    "eval_steps_per_second": 0.414
 }