isakbos_Q8BERT_COLA_L_512-finetuned-lora-tweet_eval_irony

This model is a fine-tuned version of isakbos/Q8BERT_COLA_L_512 on the tweet_eval dataset. It achieves the following results on the evaluation set:

• accuracy: 0.5686

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: 0.0005
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 8

Training results

accuracy	train_loss	epoch
0.5225	None	0
0.5225	0.6963	0
0.5361	0.6912	1
0.5560	0.6647	2
0.5476	0.6530	3
0.5518	0.6449	4
0.5518	0.6392	5
0.5613	0.6357	6
0.5686	0.6307	7

Framework versions

• PEFT 0.8.2
• Transformers 4.37.2
• Pytorch 2.2.0
• Datasets 2.16.1
• Tokenizers 0.15.2