metadata

license: mit
library_name: peft
tags:
  - alignment-handbook
  - generated_from_trainer
  - trl
  - dpo
  - generated_from_trainer
base_model: microsoft/phi-2
datasets:
  - HuggingFaceH4/ultrafeedback_binarized
model-index:
  - name: phi-2-gpo-test-longest-iter-0
    results: []

phi-2-gpo-test-longest-iter-0

This model is a fine-tuned version of lole25/phi-2-sft-ultrachat-lora on the HuggingFaceH4/ultrafeedback_binarized dataset. It achieves the following results on the evaluation set:

Loss: 0.0103
Rewards/chosen: 0.0007
Rewards/rejected: 0.0005
Rewards/accuracies: 0.5035
Rewards/margins: 0.0001
Logps/rejected: -233.4872
Logps/chosen: -256.5486
Logits/rejected: 0.8934
Logits/chosen: 0.8359

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-06
train_batch_size: 4
eval_batch_size: 4
seed: 42
distributed_type: multi-GPU
gradient_accumulation_steps: 4
total_train_batch_size: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.0101	1.6	100	0.0102	0.0004	0.0001	0.5020	0.0003	-233.5265	-256.5746	0.8922	0.8348

Framework versions

PEFT 0.7.1
Transformers 4.36.2
Pytorch 2.2.1+cu121
Datasets 2.14.6
Tokenizers 0.15.2