End of training

README.md

@@ -2,9 +2,15 @@
 license: gemma
 base_model: tanliboy/zephyr-gemma-2-9b-sft
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: zephyr-gemma-2-9b-dpo-2
   results: []
@@ -16,17 +22,17 @@ should probably proofread and complete it, then remove this comment. -->
 [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/tanliboy/huggingface/runs/dikk0994)
 # zephyr-gemma-2-9b-dpo-2
 
-This model is a fine-tuned version of [tanliboy/zephyr-gemma-2-9b-sft](https://huggingface.co/tanliboy/zephyr-gemma-2-9b-sft) on an unknown dataset.
+This model is a fine-tuned version of [tanliboy/zephyr-gemma-2-9b-sft](https://huggingface.co/tanliboy/zephyr-gemma-2-9b-sft) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5647
-- Rewards/chosen: -0.6443
-- Rewards/rejected: -1.1499
-- Rewards/accuracies: 0.6920
-- Rewards/margins: 0.5057
-- Logps/rejected: -1434.3335
-- Logps/chosen: -958.5825
-- Logits/rejected: -14.1861
-- Logits/chosen: -14.6684
+- Loss: 0.5628
+- Rewards/chosen: -0.7292
+- Rewards/rejected: -1.2825
+- Rewards/accuracies: 0.6960
+- Rewards/margins: 0.5533
+- Logps/rejected: -1566.9301
+- Logps/chosen: -1043.5624
+- Logits/rejected: -14.1720
+- Logits/chosen: -14.6638
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9965977492802931,
+    "eval_logits/chosen": -14.663813591003418,
+    "eval_logits/rejected": -14.17200756072998,
+    "eval_logps/chosen": -1043.5623779296875,
+    "eval_logps/rejected": -1566.9300537109375,
+    "eval_loss": 0.5627544522285461,
+    "eval_rewards/accuracies": 0.6959999799728394,
+    "eval_rewards/chosen": -0.7292414307594299,
+    "eval_rewards/margins": 0.5532797574996948,
+    "eval_rewards/rejected": -1.2825212478637695,
+    "eval_runtime": 148.9641,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 13.426,
+    "eval_steps_per_second": 0.839,
     "total_flos": 0.0,
     "train_loss": 0.6157421479706003,
     "train_runtime": 11733.4361,

config.json

@@ -29,6 +29,6 @@
   "sliding_window_size": 4096,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.43.1",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 256000
 }

eval_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 0.9997382884061764,
-    "eval_logits/chosen": -14.105589866638184,
-    "eval_logits/rejected": -14.292818069458008,
-    "eval_logps/chosen": -375.1571960449219,
-    "eval_logps/rejected": -407.44989013671875,
-    "eval_loss": 0.5276930332183838,
-    "eval_rewards/accuracies": 0.6880000233650208,
-    "eval_rewards/chosen": -0.6083627939224243,
-    "eval_rewards/margins": 0.6220458149909973,
-    "eval_rewards/rejected": -1.2304086685180664,
-    "eval_runtime": 148.4787,
+    "epoch": 0.9965977492802931,
+    "eval_logits/chosen": -14.663813591003418,
+    "eval_logits/rejected": -14.17200756072998,
+    "eval_logps/chosen": -1043.5623779296875,
+    "eval_logps/rejected": -1566.9300537109375,
+    "eval_loss": 0.5627544522285461,
+    "eval_rewards/accuracies": 0.6959999799728394,
+    "eval_rewards/chosen": -0.7292414307594299,
+    "eval_rewards/margins": 0.5532797574996948,
+    "eval_rewards/rejected": -1.2825212478637695,
+    "eval_runtime": 148.9641,
     "eval_samples": 2000,
-    "eval_samples_per_second": 13.47,
-    "eval_steps_per_second": 0.842
+    "eval_samples_per_second": 13.426,
+    "eval_steps_per_second": 0.839
 }

runs/Jul25_01-23-17_action-graph-trainer/events.out.tfevents.1721883328.action-graph-trainer.762006.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d04ba8b066fe7c9986abc7fb42d8f7072e5464d05e67bd4d250128eb7fc27d4e
+size 828