© 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. 57460
    Prev
    Next
    INDEX
    Explanations

    Dividing/splitting something

    np_max-act · gemini-2.0-flash

    This neuron responds to instruction cues about splitting or continuing text across multiple messages (e.g., words like “broken,” “up,” “multiple,” “messages,” “continued”).

    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
    answered
    -0.07
    reddit
    -0.07
    _estimators
    -0.06
     هر
    -0.06
     Assessment
    -0.06
     кафед
    -0.06
     حافظه
    -0.06
    wang
    -0.06
    nama
    -0.06
     inv
    -0.06
    POSITIVE LOGITS
     Malone
    0.07
     хорош
    0.06
     fleet
    0.06
    (binding
    0.06
     Scarlet
    0.06
    .Out
    0.06
    /close
    0.06
     champ
    0.06
    нити
    0.06
     MPG
    0.06
    Activations Density 0.006%

    No Known Activations