results

This model is a fine-tuned version of Linq-AI-Research/Linq-Embed-Mistral on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2908
• Accuracy: 0.9548
• F1: 0.9543

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.2148	400	0.2494	0.9220	0.9114
0.3084	0.4296	800	0.2131	0.9337	0.9289
0.226	0.6445	1200	0.2043	0.9471	0.9455
0.1898	0.8593	1600	0.1871	0.9479	0.9462
0.1673	1.0741	2000	0.2143	0.9488	0.9485
0.1673	1.2889	2400	0.2328	0.9542	0.9543
0.1287	1.5038	2800	0.2321	0.9527	0.9520
0.1216	1.7186	3200	0.1918	0.9546	0.9546
0.1303	1.9334	3600	0.1966	0.9542	0.9534
0.0937	2.1482	4000	0.2378	0.9554	0.9554
0.0937	2.3631	4400	0.2612	0.9554	0.9551
0.0637	2.5779	4800	0.2719	0.9552	0.9551
0.0544	2.7927	5200	0.2895	0.9549	0.9552

Framework versions

• PEFT 0.17.1
• Transformers 4.57.0
• Pytorch 2.4.1+cu124
• Datasets 4.1.1
• Tokenizers 0.22.1