metadata
library_name: transformers
license: llama3.1
base_model: mlfoundations-dev/stackoverflow_25000tasks_1p
tags:
- llama-factory
- full
- trl
- dpo
- llama-factory
- generated_from_trainer
model-index:
- name: simpo-stackoverflow_25000tasks_1p
results: []
simpo-stackoverflow_25000tasks_1p
This model is a fine-tuned version of mlfoundations-dev/stackoverflow_25000tasks_1p on the mlfoundations-dev/gemma2-ultrafeedback-armorm dataset. It achieves the following results on the evaluation set:
- Loss: 2.5764
- Rewards/chosen: -41.7245
- Rewards/rejected: -48.5194
- Rewards/accuracies: 0.7931
- Rewards/margins: 6.7950
- Logps/chosen: -4.1724
- Logps/rejected: -4.8519
- Logits/chosen: -1.0221
- Logits/rejected: -1.0402
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: 8e-07
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- total_eval_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/chosen | Logps/rejected | Logits/chosen | Logits/rejected |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 2.8273 | 0.9997 | 442 | 2.5764 | -41.7245 | -48.5194 | 0.7931 | 6.7950 | -4.1724 | -4.8519 | -1.0221 | -1.0402 |
Framework versions
- Transformers 4.46.1
- Pytorch 2.3.0
- Datasets 3.1.0
- Tokenizers 0.20.3