© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 12176
    Prev
    Next
    INDEX
    Explanations

    Period token

    np_max-act · gemini-2.0-flash

    The neuron detects the decimal point in numbered legal reporter citations (e.g. the “.3d” in “So.3d”).

    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
    sitemap
    -0.07
    ुग
    -0.07
     collapse
    -0.07
     quantities
    -0.06
     Волод
    -0.06
     iterating
    -0.06
    hr
    -0.06
     survival
    -0.06
     SCO
    -0.06
     rotating
    -0.06
    POSITIVE LOGITS
    nee
    0.08
    @mail
    0.07
     (*)
    0.06
    .className
    0.06
     Uploaded
    0.06
    景
    0.06
    (sin
    0.06
    rahim
    0.06
     coconut
    0.06
     Catholics
    0.06
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.