furina_eng_loss_0.0001

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

• Loss: 0.0483
• Spearman Corr: nan

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: 0.0001
• 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	1.33	200	0.0476	nan
0.047	2.66	400	0.0480	nan
0.0471	3.99	600	0.0476	nan
0.0471	5.32	800	0.0476	nan
0.047	6.64	1000	0.0476	nan
0.0471	7.97	1200	0.0476	nan
0.0471	9.3	1400	0.0471	nan
0.0471	10.63	1600	0.0467	nan
0.0468	11.96	1800	0.0473	nan
0.0468	13.29	2000	0.0485	nan
0.047	14.62	2200	0.0478	nan
0.0466	15.95	2400	0.0478	nan
0.0466	17.28	2600	0.0495	nan
0.0467	18.6	2800	0.0474	nan
0.0468	19.93	3000	0.0473	nan
0.0468	21.26	3200	0.0483	nan

Framework versions

• Transformers 4.37.2
• Pytorch 2.2.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2