AID-Neo-125M

Model description

This model was inspired by -- and finetuned on the same dataset of -- KoboldAI's GPT-Neo-125M-AID model. This was to fix a possible oversight in the original model. You could technically consider it a "retraining" of the same model using different software

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 1
eval_batch_size: 1
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 3.0