This model is a fine-tuned version of Qwen/Qwen3-0.6B on the all_julia_snippets_format_v2 dataset. It achieves the following results on the evaluation set:

Loss: 0.9086

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
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 2
total_train_batch_size: 32
total_eval_batch_size: 16
optimizer: Use adamw_torch_fused 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: 2

Training results

Training Loss	Epoch	Step	Validation Loss
1.1523	0.1871	2000	1.1229
1.0665	0.3743	4000	1.0390
1.0281	0.5614	6000	0.9989
0.9994	0.7485	8000	0.9732
0.9821	0.9357	10000	0.9552
0.9355	1.1229	12000	0.9525
0.9518	1.3100	14000	0.9447
0.9219	1.4971	16000	0.9328
0.9256	1.6843	18000	0.9232
0.9176	1.8714	20000	0.9134

Framework versions

PEFT 0.17.1
Transformers 4.56.2
Pytorch 2.8.0+cu128
Datasets 4.0.0
Tokenizers 0.22.1

CorgiPudding/Qwen3-0.6B-Julia

sft_v2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

CorgiPudding/
Qwen3-0.6B-Julia