© 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. 57785
    Prev
    Next
    INDEX
    Explanations

    incomplete or broken texts

    np_max-act · gemini-2.0-flash

    The neuron activates specifically on masked placeholder tokens (e.g., “<MASK>”) indicating blanked‐out or fill-in slots.

    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
     keyboard
    -0.06
    ”.
    -0.06
     انجمن
    -0.06
    -ton
    -0.06
    .Xtra
    -0.06
     feet
    -0.06
     Association
    -0.06
     tube
    -0.06
     larvae
    -0.06
    ]."
    -0.06
    POSITIVE LOGITS
    nám
    0.08
     Gül
    0.07
    ableOpacity
    0.07
     interstate
    0.07
    TouchableOpacity
    0.07
    θηκαν
    0.07
    oš
    0.07
     přid
    0.07
     popis
    0.07
    .Gray
    0.06
    Activations Density 0.075%

    No Known Activations