End of training

README.md

@@ -3,9 +3,16 @@ library_name: transformers
 license: llama3.1
 base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
+- tanliboy/orca_dpo_pairs
 model-index:
 - name: lambda-llama-3-8b-ipo-test
   results: []
@@ -16,17 +23,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # lambda-llama-3-8b-ipo-test
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) on the None dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) on the HuggingFaceH4/ultrafeedback_binarized and the tanliboy/orca_dpo_pairs datasets.
 It achieves the following results on the evaluation set:
-- Loss: 0.8938
-- Rewards/chosen: -0.3750
-- Rewards/rejected: -0.6076
-- Rewards/accuracies: 0.7892
-- Rewards/margins: 0.2326
-- Logps/rejected: -3.1855
-- Logps/chosen: -2.5684
-- Logits/rejected: -3.0293
-- Logits/chosen: -2.9592
+- Loss: 0.8931
+- Rewards/chosen: -0.3610
+- Rewards/rejected: -0.5883
+- Rewards/accuracies: 0.7922
+- Rewards/margins: 0.2272
+- Logps/rejected: -3.1373
+- Logps/chosen: -2.5334
+- Logits/rejected: -2.9939
+- Logits/chosen: -2.9244
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9995638901003053,
+    "eval_logits/chosen": -2.924424171447754,
+    "eval_logits/rejected": -2.9938888549804688,
+    "eval_logps/chosen": -2.5334479808807373,
+    "eval_logps/rejected": -3.1372740268707275,
+    "eval_loss": 0.8931147456169128,
+    "eval_rewards/accuracies": 0.7921686768531799,
+    "eval_rewards/chosen": -0.36102983355522156,
+    "eval_rewards/margins": 0.22724701464176178,
+    "eval_rewards/rejected": -0.5882768034934998,
+    "eval_runtime": 114.5732,
+    "eval_samples": 2643,
+    "eval_samples_per_second": 23.068,
+    "eval_steps_per_second": 0.724,
     "total_flos": 0.0,
     "train_loss": 1.0578667073141634,
     "train_runtime": 8389.7514,

config.json

@@ -34,6 +34,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 128256
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 0.9995638901003053,
+    "eval_logits/chosen": -2.924424171447754,
+    "eval_logits/rejected": -2.9938888549804688,
+    "eval_logps/chosen": -2.5334479808807373,
+    "eval_logps/rejected": -3.1372740268707275,
+    "eval_loss": 0.8931147456169128,
+    "eval_rewards/accuracies": 0.7921686768531799,
+    "eval_rewards/chosen": -0.36102983355522156,
+    "eval_rewards/margins": 0.22724701464176178,
+    "eval_rewards/rejected": -0.5882768034934998,
+    "eval_runtime": 114.5732,
+    "eval_samples": 2643,
+    "eval_samples_per_second": 23.068,
+    "eval_steps_per_second": 0.724
+}

runs/Sep21_19-14-54_action-graph-trainer/events.out.tfevents.1726955552.action-graph-trainer.3275481.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68b7e0f1d107a20f18e558261b942cb807d83a4f008dbd47a9f393c1e605cfea
+size 828