© 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. 6371
    Prev
    Next
    INDEX
    Explanations

    type

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric tokens that include decimal fractions (e.g. “2.5625,” “1.455078125,” “0.498291015625”), effectively detecting numbers with a decimal component.

    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
     chơi
    -0.07
    ABILITY
    -0.07
    AN
    -0.07
     shl
    -0.07
     вообще
    -0.06
     Sala
    -0.06
     کیل
    -0.06
    álního
    -0.06
    anus
    -0.06
    Tickets
    -0.06
    POSITIVE LOGITS
     ^{°}
    0.07
    .unlock
    0.06
     روم
    0.06
    INK
    0.06
     maxValue
    0.06
     "-";↵
    0.06
     klient
    0.06
     Category
    0.06
     popul
    0.06
    食べ
    0.06
    Activations Density 0.003%

    No Known Activations