End of training

README.md

@@ -3,10 +3,16 @@ library_name: transformers
 license: gemma
 base_model: google/gemma-7b
 tags:
+- alignment-handbook
+- trl
+- orpo
+- generated_from_trainer
 - trl
 - orpo
 - alignment-handbook
 - generated_from_trainer
+datasets:
+- silviasapora/low_quality_dpo7k
 model-index:
 - name: gemma-7b-orpo-low-quality
   results: []
@@ -17,7 +23,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gemma-7b-orpo-low-quality
 
-This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on an unknown dataset.
+This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the silviasapora/low_quality_dpo7k dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.5517
 - Rewards/chosen: -0.0554

all_results.json

@@ -1,21 +1,21 @@
 {
     "epoch": 2.9865871833084947,
-    "eval_log_odds_chosen": 0.2937372922897339,
-    "eval_log_odds_ratio": -0.6945178508758545,
-    "eval_logits/chosen": 300.6891174316406,
-    "eval_logits/rejected": 271.8756103515625,
-    "eval_logps/chosen": -1.0802680253982544,
-    "eval_logps/rejected": -1.2502641677856445,
-    "eval_loss": 1.539820671081543,
-    "eval_nll_loss": 1.4724125862121582,
-    "eval_rewards/accuracies": 0.5395683646202087,
-    "eval_rewards/chosen": -0.05401340499520302,
-    "eval_rewards/margins": 0.00849980115890503,
-    "eval_rewards/rejected": -0.06251321732997894,
-    "eval_runtime": 112.2968,
+    "eval_log_odds_chosen": 0.28559842705726624,
+    "eval_log_odds_ratio": -0.6970076560974121,
+    "eval_logits/chosen": 297.1682434082031,
+    "eval_logits/rejected": 268.0281982421875,
+    "eval_logps/chosen": -1.1085351705551147,
+    "eval_logps/rejected": -1.2919707298278809,
+    "eval_loss": 1.5517091751098633,
+    "eval_nll_loss": 1.4855411052703857,
+    "eval_rewards/accuracies": 0.5611510872840881,
+    "eval_rewards/chosen": -0.055426761507987976,
+    "eval_rewards/margins": 0.009171773679554462,
+    "eval_rewards/rejected": -0.06459853798151016,
+    "eval_runtime": 112.1993,
     "eval_samples": 553,
-    "eval_samples_per_second": 4.924,
-    "eval_steps_per_second": 1.238,
+    "eval_samples_per_second": 4.929,
+    "eval_steps_per_second": 1.239,
     "total_flos": 0.0,
     "train_loss": 1.4570662823027956,
     "train_runtime": 13599.7579,

config.json

@@ -24,6 +24,6 @@
   "rope_theta": 10000.0,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 256000
 }

eval_results.json

@@ -1,19 +1,19 @@
 {
     "epoch": 2.9865871833084947,
-    "eval_log_odds_chosen": 0.2937372922897339,
-    "eval_log_odds_ratio": -0.6945178508758545,
-    "eval_logits/chosen": 300.6891174316406,
-    "eval_logits/rejected": 271.8756103515625,
-    "eval_logps/chosen": -1.0802680253982544,
-    "eval_logps/rejected": -1.2502641677856445,
-    "eval_loss": 1.539820671081543,
-    "eval_nll_loss": 1.4724125862121582,
-    "eval_rewards/accuracies": 0.5395683646202087,
-    "eval_rewards/chosen": -0.05401340499520302,
-    "eval_rewards/margins": 0.00849980115890503,
-    "eval_rewards/rejected": -0.06251321732997894,
-    "eval_runtime": 112.2968,
+    "eval_log_odds_chosen": 0.28559842705726624,
+    "eval_log_odds_ratio": -0.6970076560974121,
+    "eval_logits/chosen": 297.1682434082031,
+    "eval_logits/rejected": 268.0281982421875,
+    "eval_logps/chosen": -1.1085351705551147,
+    "eval_logps/rejected": -1.2919707298278809,
+    "eval_loss": 1.5517091751098633,
+    "eval_nll_loss": 1.4855411052703857,
+    "eval_rewards/accuracies": 0.5611510872840881,
+    "eval_rewards/chosen": -0.055426761507987976,
+    "eval_rewards/margins": 0.009171773679554462,
+    "eval_rewards/rejected": -0.06459853798151016,
+    "eval_runtime": 112.1993,
     "eval_samples": 553,
-    "eval_samples_per_second": 4.924,
-    "eval_steps_per_second": 1.238
+    "eval_samples_per_second": 4.929,
+    "eval_steps_per_second": 1.239
 }

runs/Sep21_02-30-57_65ecb96dba42/events.out.tfevents.1726899896.65ecb96dba42.269402.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a15d570d5577cadce536887d75d2d2510944ebe8ce1d422996466359dbc4db1
+size 997