sahandrez/pairwise-reward-gemma-2-2b-it-ultrafeedback

pairwise-reward-gemma-2-2b-it_ultrafeedback_binarized_20240906_140144

This model is a fine-tuned version of google/gemma-2-2b-it on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5049
• Accuracy: 0.7519

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.483	0.0891	100	0.5471	0.7349
0.4683	0.1783	200	0.5293	0.7332
0.5305	0.2674	300	0.5220	0.7306
0.5348	0.3565	400	0.5129	0.7434
0.5165	0.4456	500	0.5090	0.7519
0.5116	0.5348	600	0.5073	0.7511
0.5209	0.6239	700	0.5076	0.7528
0.4792	0.7130	800	0.5058	0.7545
0.4474	0.8021	900	0.5047	0.7528
0.4762	0.8913	1000	0.5045	0.7511
0.5058	0.9804	1100	0.5049	0.7519

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1