© 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. 58407
    Prev
    Next
    INDEX
    Explanations

    extends

    np_max-act · gemini-2.0-flash

    The neuron is looking for exception‐related class declarations (tokens like “Exception,” “extends,” and exception class names).

    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
     nbr
    -0.06
     Nah
    -0.06
    .show
    -0.06
     LogLevel
    -0.06
    .Target
    -0.06
    posable
    -0.06
    _markup
    -0.06
    _decrypt
    -0.06
    IsUnicode
    -0.06
     منه
    -0.06
    POSITIVE LOGITS
     stub
    0.07
    whereIn
    0.07
    -esque
    0.07
    -----------
    0.07
     airstrikes
    0.07
     Recomm
    0.07
    CrLf
    0.06
    ,st
    0.06
     renowned
    0.06
     Miy
    0.06
    Activations Density 0.008%

    No Known Activations