---
library_name: peft
tags:
- alignment-handbook
- trl
- dpo
- generated_from_trainer
base_model: NbAiLab/nb-gpt-j-6B-v2
datasets:
- hugodk-sch/aftonposten_title_prefs
model-index:
- name: aftonposten-6b-align-scan
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# aftonposten-6b-align-scan

This model is a fine-tuned version of [data/ap-gpt-j-6b-sft-qlora-04-08](https://huggingface.co/data/ap-gpt-j-6b-sft-qlora-04-08) on the hugodk-sch/aftonposten_title_prefs dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3797
- Rewards/chosen: 0.0228
- Rewards/rejected: 0.0164
- Rewards/accuracies: 0.5336
- Rewards/margins: 0.0064
- Logps/rejected: -37.4984
- Logps/chosen: -34.0092
- Logits/rejected: -2.2334
- Logits/chosen: -2.2383

## 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: 8
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1

### Training results

| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
| 0.3038        | 0.26  | 100  | 0.3512          | 0.0196         | 0.0046           | 0.5424             | 0.0150          | -37.5115       | -34.0128     | -2.2324         | -2.2372       |
| 0.3157        | 0.52  | 200  | 0.3716          | 0.0148         | -0.0016          | 0.5245             | 0.0164          | -37.5184       | -34.0181     | -2.2322         | -2.2371       |
| 0.2156        | 0.78  | 300  | 0.3845          | 0.0182         | 0.0177           | 0.4934             | 0.0005          | -37.4970       | -34.0143     | -2.2316         | -2.2364       |


### Framework versions

- PEFT 0.8.2
- Transformers 4.37.2
- Pytorch 2.1.2+cu121
- Datasets 2.17.0
- Tokenizers 0.15.1