metadata
library_name: transformers
license: apache-2.0
base_model: alignment-handbook/zephyr-7b-sft-full
tags:
- alignment-handbook
- trl
- dpo
- generated_from_trainer
- trl
- dpo
- generated_from_trainer
datasets:
- HuggingFaceH4/ultrafeedback_binarized
model-index:
- name: zephyr-7b-align-scan-0.0-0.0-polynomial-3
results: []
zephyr-7b-align-scan-0.0-0.0-polynomial-3
This model is a fine-tuned version of alignment-handbook/zephyr-7b-sft-full on the HuggingFaceH4/ultrafeedback_binarized dataset. It achieves the following results on the evaluation set:
- Loss: 0.7364
- Rewards/chosen: -2.3188
- Rewards/rejected: -3.0685
- Rewards/accuracies: 0.3194
- Rewards/margins: 0.7497
- Logps/rejected: -344.3773
- Logps/chosen: -273.4246
- Logits/rejected: 1.7116
- Logits/chosen: 1.6027
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: 9.787906592614782e-07
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: polynomial
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.5825 | 1.0417 | 100 | 0.6482 | -0.2552 | -0.4894 | 0.3234 | 0.2342 | -123.1158 | -96.3836 | -2.0644 | -2.0903 |
| 0.3884 | 2.0833 | 200 | 0.6749 | -1.2868 | -1.7792 | 0.3313 | 0.4925 | -233.7709 | -184.8848 | -1.4185 | -1.4498 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.0
- Datasets 2.21.0
- Tokenizers 0.19.1