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-dpo
   results: []
@@ -15,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-gemma-dpo
 
-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 None 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: 0.4767
-- Rewards/chosen: -2.8814
-- Rewards/rejected: -4.4710
-- Rewards/accuracies: 0.7188
-- Rewards/margins: 1.5897
-- Logps/rejected: -449.6920
-- Logps/chosen: -420.1344
-- Logits/rejected: 89.6189
-- Logits/chosen: 95.6517
+- Loss: 0.4621
+- Rewards/chosen: -2.9012
+- Rewards/rejected: -4.5213
+- Rewards/accuracies: 0.7292
+- Rewards/margins: 1.6201
+- Logps/rejected: -450.6970
+- Logps/chosen: -420.5302
+- Logits/rejected: 89.5207
+- Logits/chosen: 95.5635
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.97,
+    "eval_logits/chosen": 95.56348419189453,
+    "eval_logits/rejected": 89.52069091796875,
+    "eval_logps/chosen": -420.5301818847656,
+    "eval_logps/rejected": -450.6969909667969,
+    "eval_loss": 0.462095707654953,
+    "eval_rewards/accuracies": 0.7291666865348816,
+    "eval_rewards/chosen": -2.9011573791503906,
+    "eval_rewards/margins": 1.6201168298721313,
+    "eval_rewards/rejected": -4.521275043487549,
+    "eval_runtime": 47.0999,
+    "eval_samples": 750,
+    "eval_samples_per_second": 15.924,
+    "eval_steps_per_second": 0.51,
     "train_loss": 0.38778070876231563,
     "train_runtime": 1146.4695,
     "train_samples": 6750,

config.json

@@ -23,6 +23,6 @@
   "rope_theta": 10000.0,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.38.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 256000
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 1.97,
+    "eval_logits/chosen": 95.56348419189453,
+    "eval_logits/rejected": 89.52069091796875,
+    "eval_logps/chosen": -420.5301818847656,
+    "eval_logps/rejected": -450.6969909667969,
+    "eval_loss": 0.462095707654953,
+    "eval_rewards/accuracies": 0.7291666865348816,
+    "eval_rewards/chosen": -2.9011573791503906,
+    "eval_rewards/margins": 1.6201168298721313,
+    "eval_rewards/rejected": -4.521275043487549,
+    "eval_runtime": 47.0999,
+    "eval_samples": 750,
+    "eval_samples_per_second": 15.924,
+    "eval_steps_per_second": 0.51
+}

runs/Mar05_14-57-36_node-0/events.out.tfevents.1709680871.node-0.426906.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ed54db74f5738eb2b657d99989a43f5b748bb152cc030aac7083c5c5264deeb
+size 815