Finetune_XLM_R_large_QA_New

This model is a fine-tuned version of FacebookAI/xlm-roberta-large on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.4563
• Exact: 73.0205
• F1: 84.8987
• Total: 3814
• Hasans Exact: 73.0205
• Hasans F1: 84.8987
• Hasans Total: 3814
• Best Exact: 73.0205
• Best Exact Thresh: 0.0
• Best F1: 84.8987
• Best F1 Thresh: 0.0

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: 10
• eval_batch_size: 10
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 20
• 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
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Exact	F1	Total	Hasans Exact	Hasans F1	Hasans Total	Best Exact	Best Exact Thresh	Best F1	Best F1 Thresh
1.2079	1.0	1364	1.2493	71.6308	82.6311	3814	71.6308	82.6311	3814	71.6308	0.0	82.6311	0.0
0.8693	2.0	2728	1.2005	74.2265	85.6165	3814	74.2265	85.6165	3814	74.2265	0.0	85.6165	0.0
0.6881	3.0	4092	1.2364	73.5186	84.4396	3814	73.5186	84.4396	3814	73.5186	0.0	84.4396	0.0
0.4702	4.0	5456	1.4563	73.0205	84.8987	3814	73.0205	84.8987	3814	73.0205	0.0	84.8987	0.0

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1