creation-v1

This model is a fine-tuned version of Qwen/Qwen3-0.6B on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.2783

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: 0.0001
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_ratio: 0.15
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	3	4.8549
No log	2.0	6	4.7060
No log	3.0	9	4.4868
4.3847	4.0	12	4.2240
4.3847	5.0	15	3.9478
4.3847	6.0	18	3.6598
3.1657	7.0	21	3.4035
3.1657	8.0	24	3.4113
3.1657	9.0	27	3.2417
1.6627	10.0	30	3.2186
1.6627	11.0	33	3.1794
1.6627	12.0	36	3.1747
1.6627	13.0	39	3.2023
0.9535	14.0	42	3.2163
0.9535	15.0	45	3.2625
0.9535	16.0	48	3.3318
0.7181	17.0	51	3.3887
0.7181	18.0	54	3.3620
0.7181	19.0	57	3.3598
0.498	20.0	60	3.3226
0.498	21.0	63	3.3073
0.498	22.0	66	3.2882
0.498	23.0	69	3.2865
0.4124	24.0	72	3.2841
0.4124	25.0	75	3.2844
0.4124	26.0	78	3.2786
0.4129	27.0	81	3.2787
0.4129	28.0	84	3.2782
0.4129	29.0	87	3.2785
0.4603	30.0	90	3.2783

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.7.0
• Datasets 3.5.1
• Tokenizers 0.21.1