© 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. 39066
    Prev
    Next
    INDEX
    Explanations

    three

    np_max-act · gemini-2.0-flash

    The neuron is tracking how far into the current text span you are—it steadily ramps up activation for later tokens (serving effectively as a learned positional index).

    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
     Spinner
    -0.07
    iks
    -0.07
    ematik
    -0.07
    _pat
    -0.06
    arent
    -0.06
     BY
    -0.06
     born
    -0.06
    rière
    -0.06
    預
    -0.06
    Harry
    -0.06
    POSITIVE LOGITS
    ’nın
    0.06
    子の
    0.06
    shan
    0.06
     přitom
    0.06
    .help
    0.06
     เจ
    0.06
     تاريخ
    0.06
    /history
    0.05
    .ttf
    0.05
    /year
    0.05
    Activations Density 0.029%

    No Known Activations