metadata

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-kto
    results: []

zephyr-7b-gemma-kto

This model is a fine-tuned version of HuggingFaceH4/zephyr-7b-gemma-sft-v0.1 on the argilla/dpo-mix-7k dataset. It achieves the following results on the evaluation set:

Loss: 0.2981
Rewards/chosen: 1.5381
Rewards/rejected: -0.1185
Rewards/accuracies: 0.6979
Rewards/margins: 1.6565
Logps/rejected: -364.4402
Logps/chosen: -332.9066
Logits/rejected: 106.1137
Logits/chosen: 111.3681

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: 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
0.1942	1.8957	100	0.2925	1.5810	-0.0630	0.6771	1.6440	-363.3305	-332.0488	106.0414	111.2989

Framework versions

Transformers 4.40.1
Pytorch 2.1.2+cu121
Datasets 2.19.0
Tokenizers 0.19.1