This model is a fine-tuned version of Qwen/Qwen3-4B-Instruct-2507 on the appworld_distillation_sft dataset. It achieves the following results on the evaluation set:

Loss: 0.2588

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-06
train_batch_size: 4
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 8
total_train_batch_size: 32
total_eval_batch_size: 8
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
lr_scheduler_warmup_ratio: 0.1
num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.9102	1.0	2	0.8434
0.6723	2.0	4	0.4704
0.3556	3.0	6	0.3564
0.3261	4.0	8	0.3239
0.2697	5.0	10	0.2971
0.2508	6.0	12	0.2797
0.225	7.0	14	0.2676
0.2327	8.0	16	0.2617
0.2064	9.0	18	0.2594
0.22	10.0	20	0.2588

Framework versions

Transformers 4.52.4
Pytorch 2.8.0+cu128
Datasets 3.6.0
Tokenizers 0.21.1

huseyinatahaninan/appworld_distillation_sft-SFT-Qwen3-4B-Instruct-2507

appworld_distillation_sft-SFT-Qwen3-4B-Instruct-2507

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

huseyinatahaninan/
appworld_distillation_sft-SFT-Qwen3-4B-Instruct-2507