e5_RSE_MultiLabel_08092025

This model is a fine-tuned version of intfloat/multilingual-e5-large-instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2425
• F1 Weighted: 0.8843

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: 3e-06
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Weighted
1.0829	1.0	178	0.8684	0.5980
0.8123	2.0	356	0.5814	0.6966
0.6025	3.0	534	0.4584	0.7583
0.4907	4.0	712	0.3871	0.7954
0.422	5.0	890	0.3562	0.8068
0.3762	6.0	1068	0.3218	0.8299
0.3347	7.0	1246	0.3066	0.8399
0.3032	8.0	1424	0.2843	0.8505
0.2782	9.0	1602	0.2726	0.8592
0.2527	10.0	1780	0.2639	0.8653
0.2352	11.0	1958	0.2632	0.8648
0.2211	12.0	2136	0.2549	0.8717
0.2076	13.0	2314	0.2557	0.8746
0.1963	14.0	2492	0.2467	0.8816
0.1865	15.0	2670	0.2471	0.8808
0.1808	16.0	2848	0.2471	0.8829
0.1755	17.0	3026	0.2391	0.8873
0.1723	18.0	3204	0.2404	0.8853
0.1665	19.0	3382	0.2434	0.8848
0.1643	20.0	3560	0.2425	0.8843

Framework versions

• Transformers 4.56.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.0