metadata
library_name: peft
tags:
- alignment-handbook
- trl
- dpo
- generated_from_trainer
base_model: norallm/normistral-7b-warm
datasets:
- hugodk-sch/aftonposten_title_prefs
model-index:
- name: ap-normistral-7b-align-scan
results: []
ap-normistral-7b-align-scan
This model is a fine-tuned version of data/ap-normistral-7b-sft-qlora on the hugodk-sch/aftonposten_title_prefs dataset. It achieves the following results on the evaluation set:
- Loss: 25.3941
- Rewards/chosen: -0.0157
- Rewards/rejected: -0.0327
- Rewards/accuracies: 0.5507
- Rewards/margins: 0.0170
- Logps/rejected: -36.2932
- Logps/chosen: -32.6000
- Logits/rejected: 98.6400
- Logits/chosen: 98.6695
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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 23.2882 | 0.26 | 100 | 25.6712 | -0.0029 | -0.0100 | 0.5228 | 0.0071 | -36.0669 | -32.4722 | 98.7464 | 98.7592 |
| 20.2659 | 0.52 | 200 | 24.8289 | -0.0134 | -0.0338 | 0.5341 | 0.0204 | -36.3046 | -32.5775 | 98.6263 | 98.6515 |
| 20.0695 | 0.78 | 300 | 26.1545 | -0.0226 | -0.0314 | 0.5307 | 0.0088 | -36.2802 | -32.6693 | 98.6503 | 98.6807 |
Framework versions
- PEFT 0.10.0
- Transformers 4.39.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.15.1