# tinker_cookbook.hyperparam_utils

| Function | Description |
| --- | --- |
| [`get_lora_lr_over_full_finetune_lr()`](https://tinker-docs.thinkingmachines.ai/cookbook/api-reference/hyperparam_utils/get_lora_lr_over_full_finetune_lr/) | Return the factor that you should scale the full fine-tuning learning rate by to |
| [`get_lora_param_count()`](https://tinker-docs.thinkingmachines.ai/cookbook/api-reference/hyperparam_utils/get_lora_param_count/) | Get the number of parameters in the LoRA adapter. |
| [`get_lr()`](https://tinker-docs.thinkingmachines.ai/cookbook/api-reference/hyperparam_utils/get_lr/) | Get a recommended learning rate for the given model. |
| [`get_full_finetune_param_count()`](https://tinker-docs.thinkingmachines.ai/cookbook/api-reference/hyperparam_utils/get_full_finetune_param_count/) | Get the total parameter count for a model by reading safetensors headers. |
| [`get_full_finetune_lr_multiplier()`](https://tinker-docs.thinkingmachines.ai/cookbook/api-reference/hyperparam_utils/get_full_finetune_lr_multiplier/) | Get a model-specific LR multiplier for full fine-tuning, proportional to 1/sqrt( |
| [`get_lora_lr_multiplier()`](https://tinker-docs.thinkingmachines.ai/cookbook/api-reference/hyperparam_utils/get_lora_lr_multiplier/) | Get a model-specific multiplier for the LR, when training with LoRA. |