mphd1/opt6.7b-dolly

opt6.7b

This model is a fine-tuned version of facebook/opt-6.7b on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.6760

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 1
• optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
1.8948	1.0	538	1.8861
1.1137	2.0	1076	1.9893
0.5561	3.0	1614	2.2066
0.2579	4.0	2152	2.4634
0.1481	5.0	2690	2.6760

Framework versions

• Transformers 5.14.1
• Pytorch 2.13.0+cu130
• Datasets 5.0.0
• Tokenizers 0.22.2