|
--- |
|
license: apache-2.0 |
|
base_model: martimfasantos/sft-xcomet_xl_xxl-chosen-10lp-shuff-full-tiny_new |
|
tags: |
|
- xcomet_xl_xxl |
|
- generated_from_trainer |
|
model-index: |
|
- name: dpo-hinge-xcomet_xl_xxl-10p-shuff-2e-7-full-from-sft-tiny |
|
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. --> |
|
|
|
# dpo-hinge-xcomet_xl_xxl-10p-shuff-2e-7-full-from-sft-tiny |
|
|
|
This model is a fine-tuned version of [martimfasantos/sft-xcomet_xl_xxl-chosen-10lp-shuff-full-tiny_new](https://huggingface.co/martimfasantos/sft-xcomet_xl_xxl-chosen-10lp-shuff-full-tiny_new) on the Unbabel/TowerAligned-v0.1 dataset. |
|
|
|
## 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: 2e-07 |
|
- train_batch_size: 1 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 2 |
|
- gradient_accumulation_steps: 32 |
|
- total_train_batch_size: 64 |
|
- total_eval_batch_size: 8 |
|
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 3 |
|
|
|
### Training results |
|
|
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.43.3 |
|
- Pytorch 2.3.1+cu121 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |
|
|
