tiny-audio-granite-qwen

This model is a fine-tuned version of on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3542

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: 0.001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 44
• gradient_accumulation_steps: 3
• total_train_batch_size: 96
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine_with_min_lr
• lr_scheduler_warmup_steps: 700
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.2497	0.0607	2000	0.4406
0.2300	0.1215	4000	0.4239
0.2224	0.1822	6000	0.4095
0.2128	0.2430	8000	0.3989
0.2140	0.3037	10000	0.3950
0.1957	0.3645	12000	0.3900
0.1979	0.4252	14000	0.3892
0.1877	0.4860	16000	0.3828
0.1856	0.5467	18000	0.3730
0.1794	0.6075	20000	0.3668
0.1818	0.6682	22000	0.3629
0.1696	0.7290	24000	0.3635
0.1685	0.7897	26000	0.3580
0.1645	0.8505	28000	0.3566
0.1674	0.9112	30000	0.3552
0.1643	0.9720	32000	0.3522
0.1675	1.0	32922	0.3542

Framework versions

• Transformers 5.17.0
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.23.2