© 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 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. 38787
    Prev
    Next
    INDEX
    Explanations

    multiplication

    np_max-act · gemini-2.0-flash

    The neuron activates specifically on floating-point numeric tokens (the internal probability/score numbers) in the model’s output.

    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
    410
    -0.07
    portun
    -0.07
    rary
    -0.07
    banana
    -0.07
     wooded
    -0.07
    cup
    -0.07
     Burma
    -0.07
     implants
    -0.07
     implant
    -0.06
    session
    -0.06
    POSITIVE LOGITS
    ILED
    0.06
    0.05
     alleged
    0.05
     <!--
    0.05
     Credit
    0.05
     strengthen
    0.05
    TEXT
    0.05
    .setHorizontal
    0.05
     изображ
    0.05
    (Activity
    0.05
    Activations Density 0.013%

    No Known Activations