End of training

README.md

@@ -2,9 +2,15 @@
 license: other
 base_model: HuggingFaceH4/zephyr-7b-gemma-sft-v0.1
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- argilla/dpo-mix-7k
 model-index:
 - name: zephyr-7b-gemma-kto
   results: []
@@ -15,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-gemma-kto
 
-This model is a fine-tuned version of [HuggingFaceH4/zephyr-7b-gemma-sft-v0.1](https://huggingface.co/HuggingFaceH4/zephyr-7b-gemma-sft-v0.1) on an unknown dataset.
+This model is a fine-tuned version of [HuggingFaceH4/zephyr-7b-gemma-sft-v0.1](https://huggingface.co/HuggingFaceH4/zephyr-7b-gemma-sft-v0.1) on the argilla/dpo-mix-7k dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2925
-- Rewards/chosen: 1.5810
-- Rewards/rejected: -0.0630
-- Rewards/accuracies: 0.6771
-- Rewards/margins: 1.6440
-- Logps/rejected: -363.3305
-- Logps/chosen: -332.0488
-- Logits/rejected: 106.0414
-- Logits/chosen: 111.2989
+- Loss: 0.2981
+- Rewards/chosen: 1.5381
+- Rewards/rejected: -0.1185
+- Rewards/accuracies: 0.6979
+- Rewards/margins: 1.6565
+- Logps/rejected: -364.4402
+- Logps/chosen: -332.9066
+- Logits/rejected: 106.1137
+- Logits/chosen: 111.3681
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.971563981042654,
+    "eval_logits/chosen": 111.36813354492188,
+    "eval_logits/rejected": 106.11373901367188,
+    "eval_logps/chosen": -332.9066467285156,
+    "eval_logps/rejected": -364.4402160644531,
+    "eval_loss": 0.2980845868587494,
+    "eval_rewards/accuracies": 0.6979166865348816,
+    "eval_rewards/chosen": 1.5380783081054688,
+    "eval_rewards/margins": 1.6565409898757935,
+    "eval_rewards/rejected": -0.11846256256103516,
+    "eval_runtime": 119.8229,
+    "eval_samples": 750,
+    "eval_samples_per_second": 6.259,
+    "eval_steps_per_second": 0.2,
     "total_flos": 0.0,
     "train_loss": 0.29734468632019484,
     "train_runtime": 2303.4906,

config.json

@@ -24,6 +24,6 @@
   "rope_theta": 10000.0,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.40.1",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 256000
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 1.971563981042654,
+    "eval_logits/chosen": 111.36813354492188,
+    "eval_logits/rejected": 106.11373901367188,
+    "eval_logps/chosen": -332.9066467285156,
+    "eval_logps/rejected": -364.4402160644531,
+    "eval_loss": 0.2980845868587494,
+    "eval_rewards/accuracies": 0.6979166865348816,
+    "eval_rewards/chosen": 1.5380783081054688,
+    "eval_rewards/margins": 1.6565409898757935,
+    "eval_rewards/rejected": -0.11846256256103516,
+    "eval_runtime": 119.8229,
+    "eval_samples": 750,
+    "eval_samples_per_second": 6.259,
+    "eval_steps_per_second": 0.2
+}

runs/Apr27_18-24-55_660111d13776/events.out.tfevents.1714241637.660111d13776.66089.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6aecae8e80b9dec26de3aa4eafbe476fc4c3d3ef3573c06c407fabac6748dac7
+size 815