xlm-roberta-base-MCQ-Answering-checkpoint-2

This model is a fine-tuned version of model_weights/checkpoint-15000 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.7021
• Accuracy: 0.5656

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: 16
• total_train_batch_size: 32
• optimizer: Use 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_ratio: 0.1
• training_steps: 20000

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
0.9387	0.9310	2500	0.5157	1.2488
1.0412	1.8621	5000	0.5483	1.1065
0.9125	2.7932	7500	0.5583	1.1139
0.7754	3.7239	10000	0.5547	1.1686
0.6385	4.6554	12500	0.5675	1.3830
0.5158	5.5862	15000	0.5673	1.4813
0.4455	6.5176	17500	1.6343	0.5636
0.4054	7.4484	20000	1.7021	0.5656

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0