metadata

license: other
base_model: lewtun/gemma-7b-sft-full-deita-10k-v0
tags:
  - alignment-handbook
  - trl
  - dpo
  - generated_from_trainer
  - trl
  - dpo
  - generated_from_trainer
datasets:
  - argilla/dpo-mix-7k
model-index:
  - name: gemma-7b-dpo-full-mix1-beta-0.05-epoch-3
    results: []

gemma-7b-dpo-full-mix1-beta-0.05-epoch-3

This model is a fine-tuned version of lewtun/gemma-7b-sft-full-deita-10k-v0 on the argilla/dpo-mix-7k dataset. It achieves the following results on the evaluation set:

Loss: 0.4924
Rewards/chosen: -3.4915
Rewards/rejected: -5.2635
Rewards/accuracies: 0.7396
Rewards/margins: 1.7720
Logps/rejected: -556.8203
Logps/chosen: -523.2347
Logits/rejected: 83.0993
Logits/chosen: 88.9292

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-07
train_batch_size: 2
eval_batch_size: 4
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 8
total_train_batch_size: 128
total_eval_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.1623	1.9	100	0.4861	-2.9739	-4.4977	0.6771	1.5238	-541.5043	-512.8825	87.7248	93.8502

Framework versions

Transformers 4.39.0.dev0
Pytorch 2.1.2+cu121
Datasets 2.14.6
Tokenizers 0.15.1