© 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. 54561
    Prev
    Next
    INDEX
    Explanations

    lengthy discussions

    np_max-act · gemini-2.0-flash

    The neuron activates on words that signal exhaustive or complete listings (e.g. “all,” “comprehensive,” “list”).

    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
    _boxes
    -0.06
    -AA
    -0.06
    tres
    -0.06
    .dao
    -0.06
    کو
    -0.06
     Commissioners
    -0.06
     respond
    -0.06
    spinner
    -0.05
    kip
    -0.05
    .bias
    -0.05
    POSITIVE LOGITS
    潮
    0.08
    ICATION
    0.07
     학
    0.07
    (forms
    0.07
    mentor
    0.07
     효
    0.07
     cartoon
    0.07
    点击
    0.07
    themes
    0.06
     dirección
    0.06
    Activations Density 0.197%

    No Known Activations