distilbert-base-uncased-distilled-squad_07112024T113312

This model is a fine-tuned version of distilbert/distilbert-base-uncased-distilled-squad on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8186
• F1: 0.7804
• Learning Rate: 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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Rate
No log	1.0	121	1.1959	0.5325	0.0000
No log	2.0	242	1.0212	0.6350	0.0000
No log	3.0	363	0.9424	0.6796	0.0000
No log	4.0	484	0.8570	0.7203	0.0000
1.0311	5.0	605	0.8567	0.7359	0.0000
1.0311	6.0	726	0.8271	0.7530	0.0000
1.0311	7.0	847	0.8387	0.7629	0.0000
1.0311	8.0	968	0.8491	0.7690	0.0000
0.4626	9.0	1089	0.7954	0.7809	2e-06
0.4626	10.0	1210	0.8186	0.7804	0.0

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.19.1