chakma-lm

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.5392

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• 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: 18
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
3.8855	1.0	53	3.3830
3.4777	2.0	106	3.2701
3.3303	3.0	159	3.0544
3.1507	4.0	212	3.1233
3.0755	5.0	265	2.9944
2.9621	6.0	318	2.9380
2.8887	7.0	371	2.7652
2.7405	8.0	424	2.7211
2.6984	9.0	477	2.6333
2.6866	10.0	530	2.5593
2.6092	11.0	583	2.6409
2.606	12.0	636	2.7579
2.6133	13.0	689	2.5129
2.5196	14.0	742	2.5859
2.5082	15.0	795	2.5665
2.4753	16.0	848	2.4857
2.4284	17.0	901	2.5324
2.4461	18.0	954	2.3652

Framework versions

• Transformers 4.56.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.0