2e73ed75895fcb815b43a842c9e7f071

This model is a fine-tuned version of meta-llama/Llama-3.1-8B on the contemmcm/clickbait dataset. It achieves the following results on the evaluation set:

• Loss: 0.1336
• Data Size: 1.0
• Epoch Runtime: 681.5825
• Accuracy: 0.9977
• F1 Macro: 0.9976

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	4.8993	0	15.9414	0.4948	0.4800
No log	1	650	3.6823	0.0078	23.3402	0.9767	0.9751
No log	2	1300	0.4280	0.0156	33.0142	0.9967	0.9965
No log	3	1950	3.5578	0.0312	54.0317	0.9564	0.9549
No log	4	2600	0.5344	0.0625	88.2564	0.9950	0.9947
0.1635	5	3250	2.0752	0.125	134.3912	0.6119	0.3796
0.719	6	3900	0.2065	0.25	231.2473	0.9956	0.9953
0.0406	7	4550	0.2335	0.5	435.0298	0.9952	0.9949
0.0777	8.0	5200	0.0778	1.0	775.2085	0.9977	0.9976
0.0	9.0	5850	0.0679	1.0	813.8673	0.9979	0.9978
0.0675	10.0	6500	1.1011	1.0	824.9701	0.9759	0.9743
0.0541	11.0	7150	0.0816	1.0	688.3611	0.9973	0.9972
0.0519	12.0	7800	0.1004	1.0	684.8865	0.9975	0.9974
0.0136	13.0	8450	0.1336	1.0	681.5825	0.9977	0.9976

Framework versions

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