---
library_name: peft
tags:
- alignment-handbook
- trl
- dpo
- generated_from_trainer
base_model: NorLLM-AI/NorMistral-7B
datasets:
- hugodk-sch/aftonposten_title_prefs
model-index:
- name: norllm-ai-normistral-7b-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. -->

# norllm-ai-normistral-7b-align-scan

This model is a fine-tuned version of [data/norllm-ai-normistral-7b-sft-qlora](https://huggingface.co/data/norllm-ai-normistral-7b-sft-qlora) on the hugodk-sch/aftonposten_title_prefs dataset.
It achieves the following results on the evaluation set:
- Loss: 0.9105
- Rewards/chosen: -0.0836
- Rewards/rejected: -0.1783
- Rewards/accuracies: 0.5515
- Rewards/margins: 0.0947
- Logps/rejected: -35.1438
- Logps/chosen: -31.4914
- Logits/rejected: -2.8211
- Logits/chosen: -2.8232

## 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 |
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
| 0.8436        | 0.26  | 100  | 0.9278          | -0.0021        | -0.0777          | 0.5984             | 0.0756          | -34.8925       | -31.2877     | -2.8097         | -2.8124       |
| 0.7459        | 0.52  | 200  | 0.9159          | -0.0693        | -0.1587          | 0.5835             | 0.0894          | -35.0949       | -31.4556     | -2.8227         | -2.8250       |
| 0.665         | 0.78  | 300  | 0.9202          | -0.0777        | -0.1629          | 0.5656             | 0.0852          | -35.1054       | -31.4766     | -2.8221         | -2.8242       |


### Framework versions

- PEFT 0.10.0
- Transformers 4.39.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.15.1