Qwen3-1.7B-SFT-science-5e-5 – AI Model by guangyangnlp | AlphaNeural AI

You can deploy this model and start earning money today!

guangyangnlp/
Qwen3-1.7B-SFT-science-5e-5

like

0

text-generation

generated_from_trainer

Qwen/Qwen3-1.7B

text-generation-inference

endpoints_compatible

Views

No views yet

Qwen3-1.7B-SFT-science-5e-5

This model is a fine-tuned version of Qwen/Qwen3-1.7B on the dolci_science_train dataset. It achieves the following results on the evaluation set:

Loss: 0.7408

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: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 32
total_train_batch_size: 128
optimizer: Use OptimizerNames.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_steps: 0.05
num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.9196	0.2985	230	0.8189
0.7628	0.5969	460	0.7862
0.7800	0.8954	690	0.7635
0.6428	1.1934	920	0.7614
0.6626	1.4918	1150	0.7514
0.6785	1.7903	1380	0.7408
0.5457	2.0882	1610	0.7668
0.5031	2.3867	1840	0.7642
0.5645	2.6852	2070	0.7633
0.5633	2.9837	2300	0.7632

Framework versions

Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 4.0.0
Tokenizers 0.22.2