5c6ab09a9ba84d6dd0d317691230ec31

This model is a fine-tuned version of studio-ousia/mluke-large on the dair-ai/emotion [split] dataset. It achieves the following results on the evaluation set:

• Loss: 1.5574
• Data Size: 1.0
• Epoch Runtime: 91.5914
• Accuracy: 0.3488
• F1 Macro: 0.0862

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-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro
No log	0	0	1.7964	0	3.6128	0.2092	0.0977
No log	1	500	1.7098	0.0078	4.6437	0.3488	0.0862
No log	2	1000	1.5986	0.0156	6.5640	0.3609	0.1022
No log	3	1500	1.6004	0.0312	8.7779	0.3488	0.0862
No log	4	2000	1.6071	0.0625	11.8784	0.3483	0.0861
0.0887	5	2500	1.5978	0.125	17.6378	0.2908	0.0751
1.5969	6	3000	1.5652	0.25	29.1170	0.2908	0.0751
0.2588	7	3500	1.5609	0.5	50.1012	0.3488	0.0862
1.5824	8.0	4000	1.5577	1.0	95.7221	0.3488	0.0862
1.5918	9.0	4500	1.5566	1.0	92.8645	0.3488	0.0862
1.5949	10.0	5000	1.5640	1.0	93.1318	0.2908	0.0751
1.5819	11.0	5500	1.5593	1.0	93.3355	0.3488	0.0862
1.5604	12.0	6000	1.5581	1.0	92.9769	0.3488	0.0862
1.5617	13.0	6500	1.5574	1.0	91.5914	0.3488	0.0862

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1