3c546d05d0213f52dba5d89ec2c3dd1c

This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Qwen-7B on the contemmcm/clickbait dataset. It achieves the following results on the evaluation set:

• Loss: 0.1039
• Data Size: 1.0
• Epoch Runtime: 554.4709
• Accuracy: 0.9979
• F1 Macro: 0.9978

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	9.0704	0	14.3163	0.6356	0.4486
No log	1	650	2.2307	0.0078	18.3861	0.9198	0.9177
No log	2	1300	0.5261	0.0156	34.8465	0.9900	0.9894
No log	3	1950	0.2386	0.0312	51.7685	0.9925	0.9921
No log	4	2600	0.0890	0.0625	78.7759	0.9983	0.9982
0.0362	5	3250	4.6971	0.125	124.3142	0.9630	0.9616
0.1985	6	3900	0.1307	0.25	157.2984	0.9985	0.9984
0.018	7	4550	0.0969	0.5	288.9159	0.9981	0.9980
0.0635	8.0	5200	0.1039	1.0	554.4709	0.9979	0.9978

Framework versions

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