© 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. 57625
    Prev
    Next
    INDEX
    Explanations

    produced

    np_max-act · gemini-2.0-flash

    The neuron fires on credit and attribution phrases—labels like “produced by,” “written by,” or “directed by” in metadata or category lines.

    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
    ğan
    -0.06
     bacon
    -0.06
     FLT
    -0.06
     počet
    -0.06
     fetish
    -0.06
     MCU
    -0.06
    ewriter
    -0.06
     censor
    -0.06
     BDSM
    -0.06
    iky
    -0.06
    POSITIVE LOGITS
     三
    0.07
    .getPosition
    0.07
     війни
    0.07
     accessing
    0.07
     Specifications
    0.07
    /order
    0.06
    -cover
    0.06
     llam
    0.06
     Household
    0.06
    _encoded
    0.06
    Activations Density 0.001%

    No Known Activations