End of training

README.md

@@ -2,9 +2,15 @@
 license: other
 base_model: lewtun/gemma-7b-sft-full-deita-10k-v0
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- argilla/dpo-mix-7k
 model-index:
 - name: gemma-7b-dpo-full-mix1-beta-0.1-epoch-3
   results: []
@@ -15,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gemma-7b-dpo-full-mix1-beta-0.1-epoch-3
 
-This model is a fine-tuned version of [lewtun/gemma-7b-sft-full-deita-10k-v0](https://huggingface.co/lewtun/gemma-7b-sft-full-deita-10k-v0) on the None dataset.
+This model is a fine-tuned version of [lewtun/gemma-7b-sft-full-deita-10k-v0](https://huggingface.co/lewtun/gemma-7b-sft-full-deita-10k-v0) on the argilla/dpo-mix-7k dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5052
-- Rewards/chosen: -2.9796
-- Rewards/rejected: -5.1227
-- Rewards/accuracies: 0.7708
-- Rewards/margins: 2.1431
-- Logps/rejected: -502.7766
-- Logps/chosen: -483.2003
-- Logits/rejected: 96.3785
-- Logits/chosen: 102.3770
+- Loss: 0.5433
+- Rewards/chosen: -3.1690
+- Rewards/rejected: -5.2516
+- Rewards/accuracies: 0.7188
+- Rewards/margins: 2.0826
+- Logps/rejected: -504.0652
+- Logps/chosen: -485.0945
+- Logits/rejected: 94.2728
+- Logits/chosen: 100.2497
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 2.96,
+    "eval_logits/chosen": 100.24970245361328,
+    "eval_logits/rejected": 94.2728271484375,
+    "eval_logps/chosen": -485.094482421875,
+    "eval_logps/rejected": -504.0652160644531,
+    "eval_loss": 0.5433239936828613,
+    "eval_rewards/accuracies": 0.71875,
+    "eval_rewards/chosen": -3.1689956188201904,
+    "eval_rewards/margins": 2.082571268081665,
+    "eval_rewards/rejected": -5.2515668869018555,
+    "eval_runtime": 49.158,
+    "eval_samples": 750,
+    "eval_samples_per_second": 15.257,
+    "eval_steps_per_second": 0.488,
     "train_loss": 0.27917395799587935,
     "train_runtime": 1799.8719,
     "train_samples": 6750,

config.json

@@ -23,6 +23,6 @@
   "rope_theta": 10000.0,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.39.0.dev0",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 256000
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 2.96,
+    "eval_logits/chosen": 100.24970245361328,
+    "eval_logits/rejected": 94.2728271484375,
+    "eval_logps/chosen": -485.094482421875,
+    "eval_logps/rejected": -504.0652160644531,
+    "eval_loss": 0.5433239936828613,
+    "eval_rewards/accuracies": 0.71875,
+    "eval_rewards/chosen": -3.1689956188201904,
+    "eval_rewards/margins": 2.082571268081665,
+    "eval_rewards/rejected": -5.2515668869018555,
+    "eval_runtime": 49.158,
+    "eval_samples": 750,
+    "eval_samples_per_second": 15.257,
+    "eval_steps_per_second": 0.488
+}