INDEX

Explanations

choice or freedom

np_max-act · gemini-2.0-flash

This neuron detects occurrences of the open‐ended “anything” token, flagging requests or statements that imply unlimited or unrestricted action.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

statements about capability or permission boundaries associated with the behavior of an AI.

oai_token-act-pair · gpt-4o-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rimon

-0.07

 proportion

-0.07

nces

-0.07

research

-0.07

.operator

-0.07

.attributes

-0.07

 حل

-0.07

�

-0.06

 tears

-0.06

au

-0.06

POSITIVE LOGITS

台灣

0.07

.SetActive

0.06

 prosecuted

0.06

][]

0.06

�

0.06

 सब

0.06

 уст

0.06

분

0.06

 parked

0.06

лика

0.05

Activations Density 0.003%

choice or freedom

This neuron detects occurrences of the open‐ended “anything” token, flagging requests or statements that imply unlimited or unrestricted action.

statements about capability or permission boundaries associated with the behavior of an AI.

No Comments

No Known Activations