© 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. 6408
    Prev
    Next
    INDEX
    Explanations

    equals signs

    np_max-act · gemini-2.0-flash

    This neuron targets structural markup lines indicating section or subsection headings (e.g. sequences of “=”, “–”, etc. under or over titles).

    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
     Cous
    -0.07
     Auch
    -0.06
    opes
    -0.06
    791
    -0.06
     fart
    -0.06
    克斯
    -0.06
    <Base
    -0.06
     Visitor
    -0.06
    USH
    -0.06
    mpp
    -0.06
    POSITIVE LOGITS
     comprehensive
    0.07
    (hero
    0.06
     па
    0.06
    ([
↵
    0.06
     correspondent
    0.06
     obsolete
    0.06
    .project
    0.06
     ||
↵
    0.06
     /^(
    0.06
    	cuda
    0.06
    Activations Density 0.011%

    No Known Activations