© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 92515
    Prev
    Next
    INDEX
    Explanations

    as

    np_max-act · gemini-2.0-flash

    The neuron flags “as long as …”–style conditional permission phrases (often with a following negation like “not harmful”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    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
     Env
    -0.06
    jíž
    -0.06
    Bir
    -0.06
     PW
    -0.06
    빌
    -0.06
     těl
    -0.06
     Exhib
    -0.06
     Bernie
    -0.06
    ╗
    -0.06
     captive
    -0.06
    POSITIVE LOGITS
    plus
    0.06
     conson
    0.06
    ast
    0.06
    _CLASSES
    0.06
    ullan
    0.06
    _Act
    0.06
    nap
    0.06
    _ANY
    0.06
     distance
    0.06
    ulator
    0.06
    Activations Density 0.007%

    No Known Activations