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
   results: []
@@ -17,7 +23,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gemma-7b-orpo
 
-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: 1.7559
 - Rewards/chosen: -0.0650

all_results.json

@@ -1,21 +1,21 @@
 {
     "epoch": 3.0,
-    "eval_log_odds_chosen": 0.23976314067840576,
-    "eval_log_odds_ratio": -0.6928443908691406,
-    "eval_logits/chosen": 340.5321350097656,
-    "eval_logits/rejected": 312.9670104980469,
-    "eval_logps/chosen": -1.0267834663391113,
-    "eval_logps/rejected": -1.1786600351333618,
-    "eval_loss": 1.455647587776184,
-    "eval_nll_loss": 1.4095592498779297,
-    "eval_rewards/accuracies": 0.5107913613319397,
-    "eval_rewards/chosen": -0.051339175552129745,
-    "eval_rewards/margins": 0.0075938161462545395,
-    "eval_rewards/rejected": -0.058932989835739136,
-    "eval_runtime": 90.3296,
+    "eval_log_odds_chosen": 0.3720725178718567,
+    "eval_log_odds_ratio": -0.699243426322937,
+    "eval_logits/chosen": 295.62017822265625,
+    "eval_logits/rejected": 266.0260314941406,
+    "eval_logps/chosen": -1.3004335165023804,
+    "eval_logps/rejected": -1.5281699895858765,
+    "eval_loss": 1.7558966875076294,
+    "eval_nll_loss": 1.6941194534301758,
+    "eval_rewards/accuracies": 0.597122311592102,
+    "eval_rewards/chosen": -0.06502167135477066,
+    "eval_rewards/margins": 0.011386833153665066,
+    "eval_rewards/rejected": -0.0764085054397583,
+    "eval_runtime": 90.3714,
     "eval_samples": 553,
-    "eval_samples_per_second": 6.122,
-    "eval_steps_per_second": 1.539,
+    "eval_samples_per_second": 6.119,
+    "eval_steps_per_second": 1.538,
     "total_flos": 0.0,
     "train_loss": 0.968865410152301,
     "train_runtime": 16784.8411,

config.json

@@ -24,6 +24,6 @@
   "rope_theta": 10000.0,
   "torch_dtype": "float32",
   "transformers_version": "4.44.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 256000
 }

eval_results.json

@@ -1,19 +1,19 @@
 {
-    "epoch": 1.0,
-    "eval_log_odds_chosen": 0.23976314067840576,
-    "eval_log_odds_ratio": -0.6928443908691406,
-    "eval_logits/chosen": 340.5321350097656,
-    "eval_logits/rejected": 312.9670104980469,
-    "eval_logps/chosen": -1.0267834663391113,
-    "eval_logps/rejected": -1.1786600351333618,
-    "eval_loss": 1.455647587776184,
-    "eval_nll_loss": 1.4095592498779297,
-    "eval_rewards/accuracies": 0.5107913613319397,
-    "eval_rewards/chosen": -0.051339175552129745,
-    "eval_rewards/margins": 0.0075938161462545395,
-    "eval_rewards/rejected": -0.058932989835739136,
-    "eval_runtime": 90.3296,
+    "epoch": 3.0,
+    "eval_log_odds_chosen": 0.3720725178718567,
+    "eval_log_odds_ratio": -0.699243426322937,
+    "eval_logits/chosen": 295.62017822265625,
+    "eval_logits/rejected": 266.0260314941406,
+    "eval_logps/chosen": -1.3004335165023804,
+    "eval_logps/rejected": -1.5281699895858765,
+    "eval_loss": 1.7558966875076294,
+    "eval_nll_loss": 1.6941194534301758,
+    "eval_rewards/accuracies": 0.597122311592102,
+    "eval_rewards/chosen": -0.06502167135477066,
+    "eval_rewards/margins": 0.011386833153665066,
+    "eval_rewards/rejected": -0.0764085054397583,
+    "eval_runtime": 90.3714,
     "eval_samples": 553,
-    "eval_samples_per_second": 6.122,
-    "eval_steps_per_second": 1.539
+    "eval_samples_per_second": 6.119,
+    "eval_steps_per_second": 1.538
 }

runs/Sep14_21-14-45_65ecb96dba42/events.out.tfevents.1726365868.65ecb96dba42.1985.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec77d5fdf99c43e6873c373e207e97de8605e29a17f2ae73661f2eb8667093df
+size 997