metadata
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: []
norllm-ai-normistral-7b-align-scan
This model is a fine-tuned version of 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: 6.0745
- Rewards/chosen: -0.0067
- Rewards/rejected: -0.0282
- Rewards/accuracies: 0.5660
- Rewards/margins: 0.0215
- Logps/rejected: -34.8390
- Logps/chosen: -31.3159
- Logits/rejected: -2.8030
- Logits/chosen: -2.8055
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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 5.3752 | 0.26 | 100 | 6.0492 | 0.0006 | -0.0242 | 0.5893 | 0.0248 | -34.8191 | -31.2794 | -2.8110 | -2.8137 |
| 5.6214 | 0.52 | 200 | 5.9543 | -0.0060 | -0.0319 | 0.5685 | 0.0259 | -34.8574 | -31.3123 | -2.8103 | -2.8130 |
| 4.7461 | 0.78 | 300 | 5.9715 | 0.0004 | -0.0246 | 0.5748 | 0.0250 | -34.8210 | -31.2804 | -2.8025 | -2.8048 |
Framework versions
- PEFT 0.10.0
- Transformers 4.39.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.15.1