End of training

README.md

@@ -2,9 +2,15 @@
 license: mit
 base_model: microsoft/Phi-3-mini-128k-instruct
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- princeton-nlp/llama3-ultrafeedback
 model-index:
 - name: phi-3-mini-128k-instruct-dpo-lr-5e-07
   results: []
@@ -15,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # phi-3-mini-128k-instruct-dpo-lr-5e-07
 
-This model is a fine-tuned version of [microsoft/Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) on an unknown dataset.
+This model is a fine-tuned version of [microsoft/Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) on the princeton-nlp/llama3-ultrafeedback dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6104
-- Rewards/chosen: -1.0659
-- Rewards/rejected: -1.4533
+- Loss: 0.6096
+- Rewards/chosen: -1.0852
+- Rewards/rejected: -1.4834
 - Rewards/accuracies: 0.6976
-- Rewards/margins: 0.3875
-- Logps/rejected: -433.6641
-- Logps/chosen: -403.0910
-- Logits/rejected: 1.6821
-- Logits/chosen: 1.6709
+- Rewards/margins: 0.3982
+- Logps/rejected: -434.2651
+- Logps/chosen: -403.4777
+- Logits/rejected: 1.6861
+- Logits/chosen: 1.6753
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9981298423724285,
+    "eval_logits/chosen": 1.675267219543457,
+    "eval_logits/rejected": 1.6861435174942017,
+    "eval_logps/chosen": -403.47772216796875,
+    "eval_logps/rejected": -434.26507568359375,
+    "eval_loss": 0.6095927357673645,
+    "eval_rewards/accuracies": 0.6975806355476379,
+    "eval_rewards/chosen": -1.0852103233337402,
+    "eval_rewards/margins": 0.3981803059577942,
+    "eval_rewards/rejected": -1.4833906888961792,
+    "eval_runtime": 88.8294,
+    "eval_samples": 1961,
+    "eval_samples_per_second": 22.076,
+    "eval_steps_per_second": 0.698,
     "total_flos": 0.0,
     "train_loss": 0.6216196353991996,
     "train_runtime": 6954.5059,

config.json

@@ -135,6 +135,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.41.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32064
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 0.9981298423724285,
+    "eval_logits/chosen": 1.675267219543457,
+    "eval_logits/rejected": 1.6861435174942017,
+    "eval_logps/chosen": -403.47772216796875,
+    "eval_logps/rejected": -434.26507568359375,
+    "eval_loss": 0.6095927357673645,
+    "eval_rewards/accuracies": 0.6975806355476379,
+    "eval_rewards/chosen": -1.0852103233337402,
+    "eval_rewards/margins": 0.3981803059577942,
+    "eval_rewards/rejected": -1.4833906888961792,
+    "eval_runtime": 88.8294,
+    "eval_samples": 1961,
+    "eval_samples_per_second": 22.076,
+    "eval_steps_per_second": 0.698
+}

runs/Jun21_18-21-32_131-186-38-174/events.out.tfevents.1719001929.131-186-38-174.31109.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0aa11615aa27f30d81c04959eff63c616c0f81ef9563de788cab572cce02455c
+size 828