© 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. 47388
    Prev
    Next
    INDEX
    Explanations

    been

    np_max-act · gemini-2.0-flash

    It detects first-person present-perfect continuous phrases like "I've been" / "I have been" (i.e., statements about ongoing or recent personal activity).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
    Wars
    -0.06
    field
    -0.06
    raph
    -0.06
    ophil
    -0.06
    idunt
    -0.06
    entially
    -0.06
    Overlap
    -0.06
    .herokuapp
    -0.06
     manžel
    -0.06
    _debug
    -0.06
    POSITIVE LOGITS
    ประเทศไทย
    0.07
     knocks
    0.07
     vampire
    0.06
     vex
    0.06
    begin
    0.06
     tez
    0.06
     waist
    0.06
     huy
    0.06
     двиг
    0.06
     divisor
    0.06
    Activations Density 0.025%

    No Known Activations