afrolid_mega

This model is a fine-tuned version of UBC-NLP/serengeti on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0886
• F1: 0.9755

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-05
• train_batch_size: 512
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 4
• total_train_batch_size: 8192
• total_eval_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 25.0

Training results

Training Loss	Epoch	Step	Validation Loss	F1
0.0227	17.7936	5000	0.0886	0.9755

Framework versions

• Transformers 4.57.1
• Pytorch 2.11.0
• Datasets 3.6.0
• Tokenizers 0.22.2