© 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. 128261
    Prev
    Next
    INDEX
    Explanations

    example

    np_max-act · gemini-2.0-flash

    instructional phrases related to programming and coding tasks.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron fires on disclaimer/boilerplate terms (e.g. “example,” “just,” “note,” “an”) used to hedge or qualify instructions.

    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
    @api
    -0.07
    사업
    -0.07
     Fridays
    -0.07
     graphics
    -0.07
    nam
    -0.06
    _priority
    -0.06
    sandbox
    -0.06
     Newman
    -0.06
     acronym
    -0.06
    .nextSibling
    -0.06
    POSITIVE LOGITS
    .Prop
    0.06
     blanket
    0.06
     Debt
    0.06
    acers
    0.06
    light
    0.06
    Printf
    0.06
    oten
    0.06
     야
    0.06
     forms
    0.06
    138
    0.06
    Activations Density 0.021%

    No Known Activations