© 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. 58537
    Prev
    Next
    INDEX
    Explanations

    academic texts

    np_max-act · gemini-2.0-flash

    The neuron responds to quantitative or measurement-oriented tokens—section headings, numeric values, and units—marking passages that convey data, metrics, or structured “how-to-measure” content.

    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
     lact
    -0.06
    amed
    -0.06
     رمز
    -0.06
    ftware
    -0.06
    avatars
    -0.06
     стоя
    -0.06
     tatto
    -0.06
     =======
    -0.06
     hatch
    -0.06
     heap
    -0.06
    POSITIVE LOGITS
    سر
    0.07
     nghiên
    0.06
     Semiconductor
    0.06
    刑
    0.06
     Smarty
    0.06
     lối
    0.06
    >',
    0.06
     вкус
    0.06
    <!--
    0.06
    .AppCompatActivity
    0.06
    Activations Density 0.091%

    No Known Activations