loha_fine_tuned_rte_croslo

This model is a fine-tuned version of EMBEDDIA/crosloengual-bert on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6891
• Accuracy: 0.5862
• F1: 0.5862

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
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 400

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.7088	1.7241	50	0.6839	0.6552	0.6388
0.695	3.4483	100	0.6840	0.6552	0.6388
0.7126	5.1724	150	0.6869	0.5862	0.5789
0.7051	6.8966	200	0.6888	0.5862	0.5862
0.6905	8.6207	250	0.6889	0.5862	0.5862
0.7075	10.3448	300	0.6894	0.5862	0.5862
0.701	12.0690	350	0.6890	0.5862	0.5862
0.7036	13.7931	400	0.6891	0.5862	0.5862

Framework versions

• PEFT 0.11.1
• Transformers 4.40.2
• Pytorch 2.1.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1