Conventional LLMs may suffer from corpus heterogeneity and subtle condition changes. This work activates the meta-signal of $\beta$ within the SwiGLU blocks, resulting in a meta-gating mechanism that adaptively adjusts the nonlinearity of FFN. A hypernetwork is employed which dynamically produces $\beta$ on textual conditions, providing meta-controllability on LLMs.

This checkpoint is trained on a subset (non_nli_to_nli) of CrossFit and UnifiedQA.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0001
train_batch_size: 1
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 32
total_train_batch_size: 32
total_eval_batch_size: 256
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: 1.0

Framework versions

Transformers 4.46.1
Pytorch 2.3.0+cu121
Datasets 2.20.0
Tokenizers 0.20.3

Citation

If you find this work useful, please consider citing:

bibtex

@article{ji2026learntolearn,
  title={Learn-to-learn on Arbitrary Textual Conditioning: A Hypernetwork-Driven Meta-Gated LLM},
  author={Luo Ji and Qi Qin and Ningyuan Xi and Teng Chen and Qingqing Gu and Hongyan Li},
  journal={arXiv preprint arXiv:2605.01973},
  year={2026}
}

jiluoaaron/MeGan-R512-CrossFit_UnifiedQA-nn2n-lr1e-6

MeGan (meta_swiglu) - CrossFit & UnifiedQA SFT

Model description

Training procedure

Training hyperparameters

Framework versions

Citation

jiluoaaron/
MeGan-R512-CrossFit_UnifiedQA-nn2n-lr1e-6