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:
+- argilla/dpo-mix-7k
 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 argilla/dpo-mix-7k dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.6395
 - Rewards/chosen: -0.0601

all_results.json

@@ -1,21 +1,21 @@
 {
     "epoch": 2.9928514694201747,
-    "eval_log_odds_chosen": 0.26997339725494385,
-    "eval_log_odds_ratio": -0.7197813391685486,
-    "eval_logits/chosen": 284.18389892578125,
-    "eval_logits/rejected": 254.83587646484375,
-    "eval_logps/chosen": -1.2543175220489502,
-    "eval_logps/rejected": -1.4255971908569336,
-    "eval_loss": 1.7139935493469238,
-    "eval_nll_loss": 1.6444751024246216,
-    "eval_rewards/accuracies": 0.5467625856399536,
-    "eval_rewards/chosen": -0.06271587312221527,
-    "eval_rewards/margins": 0.008563979528844357,
-    "eval_rewards/rejected": -0.0712798535823822,
-    "eval_runtime": 25.9538,
+    "eval_log_odds_chosen": 0.44380733370780945,
+    "eval_log_odds_ratio": -0.6702221632003784,
+    "eval_logits/chosen": 286.3763122558594,
+    "eval_logits/rejected": 275.9735412597656,
+    "eval_logps/chosen": -1.2025552988052368,
+    "eval_logps/rejected": -1.5090675354003906,
+    "eval_loss": 1.639459252357483,
+    "eval_nll_loss": 1.5846672058105469,
+    "eval_rewards/accuracies": 0.6028881072998047,
+    "eval_rewards/chosen": -0.060127776116132736,
+    "eval_rewards/margins": 0.015325604937970638,
+    "eval_rewards/rejected": -0.07545337826013565,
+    "eval_runtime": 278.5267,
     "eval_samples": 553,
-    "eval_samples_per_second": 21.307,
-    "eval_steps_per_second": 5.356,
+    "eval_samples_per_second": 1.985,
+    "eval_steps_per_second": 0.995,
     "total_flos": 0.0,
     "train_loss": 1.4771008792703066,
     "train_runtime": 40012.5124,

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": 3.0,
-    "eval_log_odds_chosen": 0.26997339725494385,
-    "eval_log_odds_ratio": -0.7197813391685486,
-    "eval_logits/chosen": 284.18389892578125,
-    "eval_logits/rejected": 254.83587646484375,
-    "eval_logps/chosen": -1.2543175220489502,
-    "eval_logps/rejected": -1.4255971908569336,
-    "eval_loss": 1.7139935493469238,
-    "eval_nll_loss": 1.6444751024246216,
-    "eval_rewards/accuracies": 0.5467625856399536,
-    "eval_rewards/chosen": -0.06271587312221527,
-    "eval_rewards/margins": 0.008563979528844357,
-    "eval_rewards/rejected": -0.0712798535823822,
-    "eval_runtime": 25.9538,
+    "epoch": 2.9928514694201747,
+    "eval_log_odds_chosen": 0.44380733370780945,
+    "eval_log_odds_ratio": -0.6702221632003784,
+    "eval_logits/chosen": 286.3763122558594,
+    "eval_logits/rejected": 275.9735412597656,
+    "eval_logps/chosen": -1.2025552988052368,
+    "eval_logps/rejected": -1.5090675354003906,
+    "eval_loss": 1.639459252357483,
+    "eval_nll_loss": 1.5846672058105469,
+    "eval_rewards/accuracies": 0.6028881072998047,
+    "eval_rewards/chosen": -0.060127776116132736,
+    "eval_rewards/margins": 0.015325604937970638,
+    "eval_rewards/rejected": -0.07545337826013565,
+    "eval_runtime": 278.5267,
     "eval_samples": 553,
-    "eval_samples_per_second": 21.307,
-    "eval_steps_per_second": 5.356
+    "eval_samples_per_second": 1.985,
+    "eval_steps_per_second": 0.995
 }

runs/Sep17_17-52-37_f5c89788b4e2/events.out.tfevents.1726636616.f5c89788b4e2.2200488.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af1325f2ad1715f3e4f9aaeef2dddd5f3b49d4ab8f43b3253c9e2d63b95ccbdf
+size 997