distilbert-base-uncased-finetuned-sst-2-english_08112024T110433

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: 0.4561
• F1: 0.8591
• 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: 1e-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: 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	1.8025	0.1776	0.0000
No log	2.0	173	1.7296	0.1902	0.0000
No log	2.9942	259	1.6400	0.3649	0.0000
No log	4.0	346	1.4873	0.4157	0.0000
No log	4.9942	432	1.2812	0.5119	0.0000
1.5846	6.0	519	1.0792	0.5668	0.0000
1.5846	6.9942	605	0.9450	0.6226	1e-05
1.5846	8.0	692	0.8238	0.6836	0.0000
1.5846	8.9942	778	0.7274	0.7374	0.0000
1.5846	10.0	865	0.6411	0.7734	0.0000
1.5846	10.9942	951	0.6020	0.7822	0.0000
0.7489	12.0	1038	0.5470	0.8157	0.0000
0.7489	12.9942	1124	0.5161	0.8299	0.0000
0.7489	14.0	1211	0.4952	0.8454	0.0000
0.7489	14.9942	1297	0.4768	0.8518	0.0000
0.7489	16.0	1384	0.4664	0.8546	0.0000
0.7489	16.9942	1470	0.4665	0.8565	0.0000
0.3329	18.0	1557	0.4561	0.8591	5e-07
0.3329	18.9942	1643	0.4578	0.8585	1e-07
0.3329	19.8844	1720	0.4579	0.8584	0.0

Framework versions

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