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: 0.5060
- Rewards/chosen: 0.0978
- Rewards/rejected: 0.1310
- Rewards/accuracies: 0.5220
- Rewards/margins: -0.0333
- Logps/rejected: -35.7793
- Logps/chosen: -32.3035
- Logits/rejected: 98.3593
- Logits/chosen: 98.3666
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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.461 | 0.26 | 100 | 0.5000 | 0.0677 | 0.0660 | 0.5158 | 0.0017 | -35.8723 | -32.3464 | 98.6883 | 98.7043 |
| 0.3172 | 0.52 | 200 | 0.4956 | 0.0869 | 0.0578 | 0.5133 | 0.0291 | -35.8840 | -32.3190 | 98.4409 | 98.4463 |
| 0.3303 | 0.78 | 300 | 0.4980 | 0.1165 | 0.1232 | 0.5017 | -0.0067 | -35.7905 | -32.2767 | 98.3630 | 98.3713 |
Framework versions
- PEFT 0.10.0
- Transformers 4.39.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.15.1