InkubaLM-0.4B_Kumshe_English-Chinese-Hausa_Dataset_test_size_0.0008

This model is a fine-tuned version of lelapa/InkubaLM-0.4B on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.6223
• Chrf: 0.2267

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-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• 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: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Chrf
No log	1.0	71	3.8868	0.2218
No log	2.0	142	3.5556	0.2314
No log	3.0	213	3.6223	0.2267

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.1
• Tokenizers 0.21.0