© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 59061
    Prev
    Next
    INDEX
    Explanations

    Di

    np_max-act · gemini-2.0-flash

    The neuron chiefly activates on the token “Di,” i.e. the beginning of words that start with “Di.”

    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
     Powers
    -0.07
    errick
    -0.07
     Kahn
    -0.07
    <W
    -0.07
    Russ
    -0.06
    .ResultSet
    -0.06
    emm
    -0.06
     ผล
    -0.06
     Hull
    -0.06
     Burton
    -0.06
    POSITIVE LOGITS
    di
    0.12
     di
    0.12
     Di
    0.11
     Diary
    0.10
    .Di
    0.09
    оди
    0.09
    _di
    0.09
    -di
    0.09
    Di
    0.09
     días
    0.09
    Activations Density 0.018%

    No Known Activations