End of training

README.md

@@ -3,9 +3,15 @@ library_name: transformers
 license: gemma
 base_model: google/gemma-2-9b-it
 tags:
+- alignment-handbook
 - trl
 - simpo
 - generated_from_trainer
+- trl
+- simpo
+- generated_from_trainer
+datasets:
+- jz666/gemma2-ultrafeedback-ppl-split
 model-index:
 - name: gemma-2-9b-it-dpo-train_filtered_full
   results: []
@@ -16,17 +22,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gemma-2-9b-it-dpo-train_filtered_full
 
-This model is a fine-tuned version of [google/gemma-2-9b-it](https://huggingface.co/google/gemma-2-9b-it) on an unknown dataset.
+This model is a fine-tuned version of [google/gemma-2-9b-it](https://huggingface.co/google/gemma-2-9b-it) on the jz666/gemma2-ultrafeedback-ppl-split dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6509
-- Rewards/chosen: -0.1749
-- Rewards/rejected: -0.2671
-- Rewards/accuracies: 0.7561
-- Rewards/margins: 0.0922
-- Logps/rejected: -26.7055
-- Logps/chosen: -17.4891
-- Logits/rejected: -24.9311
-- Logits/chosen: -23.9323
+- Loss: 0.6501
+- Rewards/chosen: -0.1825
+- Rewards/rejected: -0.2772
+- Rewards/accuracies: 0.75
+- Rewards/margins: 0.0947
+- Logps/rejected: -27.7159
+- Logps/chosen: -18.2499
+- Logits/rejected: -25.0650
+- Logits/chosen: -24.0431
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9989094874591058,
+    "eval_logits/chosen": -24.043106079101562,
+    "eval_logits/rejected": -25.065006256103516,
+    "eval_logps/chosen": -18.249919891357422,
+    "eval_logps/rejected": -27.7159366607666,
+    "eval_loss": 0.6501222848892212,
+    "eval_rewards/accuracies": 0.75,
+    "eval_rewards/chosen": -0.18249918520450592,
+    "eval_rewards/margins": 0.09466017782688141,
+    "eval_rewards/rejected": -0.27715936303138733,
+    "eval_runtime": 86.4169,
+    "eval_samples": 1941,
+    "eval_samples_per_second": 22.461,
+    "eval_steps_per_second": 1.412,
     "total_flos": 0.0,
     "train_loss": 0.6719389883191305,
     "train_runtime": 8958.0836,

config.json

@@ -29,6 +29,6 @@
   "sliding_window_size": 4096,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 256000
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 0.9989094874591058,
+    "eval_logits/chosen": -24.043106079101562,
+    "eval_logits/rejected": -25.065006256103516,
+    "eval_logps/chosen": -18.249919891357422,
+    "eval_logps/rejected": -27.7159366607666,
+    "eval_loss": 0.6501222848892212,
+    "eval_rewards/accuracies": 0.75,
+    "eval_rewards/chosen": -0.18249918520450592,
+    "eval_rewards/margins": 0.09466017782688141,
+    "eval_rewards/rejected": -0.27715936303138733,
+    "eval_runtime": 86.4169,
+    "eval_samples": 1941,
+    "eval_samples_per_second": 22.461,
+    "eval_steps_per_second": 1.412
+}