furina_seed42_eng_esp_hau_basic_5e-06

This model is a fine-tuned version of yihongLiu/furina on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0249
• Spearman Corr: 0.7476

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: 5e-06
• train_batch_size: 32
• eval_batch_size: 128
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Spearman Corr
No log	1.45	200	0.0579	0.1807
0.1318	2.91	400	0.0365	0.5375
0.0455	4.36	600	0.0280	0.6335
0.0455	5.82	800	0.0251	0.6685
0.0329	7.27	1000	0.0255	0.6937
0.0273	8.73	1200	0.0238	0.7208
0.0247	10.18	1400	0.0232	0.7297
0.0247	11.64	1600	0.0238	0.7338
0.0229	13.09	1800	0.0232	0.7352
0.0214	14.55	2000	0.0237	0.7407
0.0204	16.0	2200	0.0246	0.7432
0.0204	17.45	2400	0.0253	0.7453
0.0191	18.91	2600	0.0254	0.7465
0.0181	20.36	2800	0.0256	0.7475
0.0181	21.82	3000	0.0249	0.7476

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2