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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 58509
    Prev
    Next
    INDEX
    Explanations

    answers and solutions

    np_max-act · gemini-2.0-flash

    The neuron spots the system instruction line that mentions a descriptive “answer” followed by a numeric score (e.g. “answer 5.28515625”).

    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
    ora
    -0.07
    ド
    -0.07
    ilerinin
    -0.07
     Linux
    -0.07
    .em
    -0.07
    -Fi
    -0.06
    .N
    -0.06
    िब
    -0.06
    يلة
    -0.06
     zároveň
    -0.06
    POSITIVE LOGITS
    �
    0.07
    .parent
    0.07
    _references
    0.06
     explain
    0.06
     있어
    0.06
    ished
    0.06
    REPORT
    0.06
     disobed
    0.06
    =target
    0.06
    昨
    0.06
    Activations Density 0.016%

    No Known Activations