© 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. 38976
    Prev
    Next
    INDEX
    Explanations

    Multimedia content

    np_max-act · gemini-2.0-flash

    The neuron is looking for the transcript’s narrator‐style commentary cues (e.g. “I see,” “I hear someone,” timestamps), i.e. the descriptive observation lines rather than the movie dialogue itself.

    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
    Fine
    -0.07
     Judges
    -0.06
     Pixels
    -0.06
    /pkg
    -0.06
    Vin
    -0.06
    DTV
    -0.06
    sdk
    -0.06
     Telescope
    -0.06
    ोव
    -0.06
    .CodeAnalysis
    -0.06
    POSITIVE LOGITS
    enqueue
    0.08
    /yyyy
    0.06
     Seeder
    0.06
    bsub
    0.06
    กระท
    0.06
    Dispatch
    0.06
    ![↵
    0.06
    Editing
    0.06
    ený
    0.06
     eagerly
    0.06
    Activations Density 0.008%

    No Known Activations