© 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. 58974
    Prev
    Next
    INDEX
    Explanations

    hobbies

    np_max-act · gemini-2.0-flash

    The neuron fires on the key single‐word labels that start list items (e.g. “Desporto,” “Dieta,” “Cultura,” etc.) in Portuguese bulleted lists.

    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
    ,default
    -0.07
    kuk
    -0.07
     junk
    -0.06
     زن
    -0.06
    _results
    -0.06
     Celt
    -0.06
    .dp
    -0.06
     рук
    -0.06
     Tar
    -0.06
     Palo
    -0.06
    POSITIVE LOGITS
    _CAPACITY
    0.06
    ैं,
    0.06
    经过
    0.06
    .adapters
    0.06
     './
    0.06
     improves
    0.06
    NYSE
    0.06
    plemented
    0.06
    -readable
    0.06
     เพ
    0.06
    Activations Density 0.156%

    No Known Activations