bge_large_cn_new_prompt_llama3_70_2

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.9337
• Precision: 0.4748
• Recall: 0.3645
• F1 Macro: 0.3499
• Accuracy: 0.4695

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: 0.0003
• train_batch_size: 2048
• eval_batch_size: 1024
• seed: 0
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1 Macro	Accuracy
No log	0	0	10.0773	0.0731	0.1667	0.0201	0.0638
0.9787	2.9326	1000	0.9896	0.4564	0.3443	0.3343	0.4531
0.9384	5.8651	2000	0.9622	0.4761	0.3558	0.3436	0.4648
0.9347	8.7977	3000	0.9476	0.4705	0.3592	0.3447	0.4594
0.9172	11.7302	4000	0.9409	0.4716	0.3608	0.3453	0.4613
0.9073	14.6628	5000	0.9369	0.4661	0.3665	0.3535	0.4648
0.8973	17.5953	6000	0.9337	0.4748	0.3645	0.3499	0.4695

Framework versions

• Transformers 4.43.3
• Pytorch 2.4.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1