© 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. 88923
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens that name an application’s entry or startup point (e.g. “ENTRY,” “Main-Class,” “StartPage,” “Page,” etc.).

    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
    ژ
    -0.07
     nin
    -0.07
     validations
    -0.06
     horizontally
    -0.06
    foundland
    -0.06
     onto
    -0.06
    ylon
    -0.06
     ly
    -0.06
    liği
    -0.06
     Ele
    -0.06
    POSITIVE LOGITS
     генера
    0.07
     %.
    0.07
     мир
    0.07
    _ONCE
    0.06
     LOGIN
    0.06
     inaug
    0.06
    OLID
    0.06
     Spec
    0.06
    _ET
    0.06
    Unity
    0.06
    Activations Density 0.023%

    No Known Activations