tdhook.latent.steering_vectors#
Classes#
Steering vectors [24]. |
|
Factory for creating hooking contexts. |
Module Contents#
- class tdhook.latent.steering_vectors.SteeringVectors(modules_to_steer, steer_fn)[source]#
Bases:
tdhook.contexts.HookingContextFactorySteering vectors [24].
- Parameters:
modules_to_steer (List[str])
steer_fn (Callable)
- _hook_module(module)[source]#
- Parameters:
module (tdhook.modules.HookedModule)
- Return type:
- class tdhook.latent.steering_vectors.ActivationAddition(modules_to_steer, positive_key='positive', negative_key='negative', steer_key='steer', clean_intermediate_keys=True, cache_callback=None)[source]#
Bases:
tdhook.contexts.HookingContextFactoryFactory for creating hooking contexts.
- Parameters:
modules_to_steer (List[str])
positive_key (tdhook._types.UnraveledKey)
negative_key (tdhook._types.UnraveledKey)
steer_key (tdhook._types.UnraveledKey)
clean_intermediate_keys (bool)
cache_callback (Optional[Callable])
- _prepare_module(module, in_keys, out_keys, extra_relative_path)[source]#
- Parameters:
module (tensordict.nn.TensorDictModuleBase)
in_keys (List[tdhook._types.UnraveledKey])
out_keys (List[tdhook._types.UnraveledKey])
extra_relative_path (str)
- Return type:
tensordict.nn.TensorDictModuleBase
- _hook_module(module)[source]#
- Parameters:
module (tdhook.modules.HookedModule)
- Return type: