© 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. 56368
    Prev
    Next
    INDEX
    Explanations

    beginning of stories

    np_max-act · gemini-2.0-flash

    The neuron activates on chunks that describe a character’s immediate setting and current state or actions (e.g., “You’re in bed,” “NAME_1 is at the pub,” “You’re bored and feeling horny,” “You’re about to message me…”).

    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
     reshape
    -0.07
     Burk
    -0.06
    舌
    -0.06
    .Pin
    -0.06
     Israeli
    -0.06
     Fiber
    -0.06
     Skinner
    -0.06
     broth
    -0.06
    resas
    -0.06
     Kelly
    -0.06
    POSITIVE LOGITS
     Concern
    0.09
    _packages
    0.06
     messed
    0.06
    .student
    0.06
    {@
    0.06
    .ali
    0.06
     milyon
    0.06
    환경
    0.06
    rient
    0.06
     <%=
    0.06
    Activations Density 0.087%

    No Known Activations