# tinker.types.LoraConfig

## _class_ [**tinker.types.LoraConfig**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/lora_config.py#L8)( _[StrictBase](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/_models.py#L76)_)

**Fields:**

- [**rank**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/lora_config.py#L9) ( _int_) – LoRA rank (dimension of low-rank matrices)
- [**seed**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/lora_config.py#L12) ( _Optional[int]_) – Seed used for initialization of LoRA weights.

Useful if you need deterministic or reproducible initialization of weights.

- [**train_unembed**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/lora_config.py#L18) ( _bool_) – Whether to add lora to the unembedding layer

- [**train_mlp**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/lora_config.py#L21) ( _bool_) – Whether to add loras to the MLP layers (including MoE layers)
- [**train_attn**](https://github.com/thinking-machines-lab/tinker/blob/main/src/tinker/types/lora_config.py#L24) ( _bool_) – Whether to add loras to the attention layers