distilbert-base-uncased-finetuned-sst-2-english_08112024T121529

This model is a fine-tuned version of distilbert/distilbert-base-uncased-finetuned-sst-2-english on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0521
• F1: 0.5770
• 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: 0.001
• 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: cosine
• lr_scheduler_warmup_steps: 600
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Rate
No log	0.9942	86	1.0521	0.5770	0.0001
No log	2.0	173	1.0938	0.6023	0.0003
No log	2.9942	259	1.1901	0.6153	0.0004
No log	4.0	346	1.1094	0.6056	0.0006
No log	4.9942	432	1.2608	0.4934	0.0007
1.1523	6.0	519	1.2942	0.5451	0.0009
1.1523	6.9942	605	1.2737	0.5362	0.001
1.1523	8.0	692	1.2844	0.4450	0.0010
1.1523	8.9942	778	1.6545	0.2937	0.0009
1.1523	10.0	865	1.4726	0.3773	0.0009
1.1523	10.9942	951	1.5591	0.3796	0.0008
1.367	12.0	1038	1.4358	0.4459	0.0007
1.367	12.9942	1124	1.4144	0.4272	0.0006
1.367	14.0	1211	1.3512	0.4938	0.0004
1.367	14.9942	1297	1.2704	0.5067	0.0003
1.367	16.0	1384	1.2057	0.5329	0.0002
1.367	16.9942	1470	1.1236	0.5658	0.0001
1.206	18.0	1557	1.1138	0.5826	0.0001
1.206	18.9942	1643	1.0703	0.6122	0.0000
1.206	19.8844	1720	1.0771	0.6069	0.0

Framework versions

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