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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 41458
    Prev
    Next
    INDEX
    Explanations

    numbers and measurements

    np_max-act · gemini-2.0-flash

    The neuron activates on the fractional (decimal) parts of numerical values, i.e. tokens representing the digits after a decimal point in measurements.

    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
     eve
    -0.06
     Bee
    -0.06
    unused
    -0.06
     =(
    -0.06
     cliffs
    -0.06
     bathing
    -0.06
    peror
    -0.06
     daughter
    -0.06
    tems
    -0.06
     usage
    -0.06
    POSITIVE LOGITS
    Enterprise
    0.06
    айт
    0.06
    _RESOURCES
    0.06
    品牌
    0.06
    _REALTYPE
    0.06
    .GetChild
    0.06
    ディア
    0.06
    =yes
    0.06
    ندي
    0.06
    財
    0.06
    Activations Density 0.147%

    No Known Activations