© 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 EvalsExportsinterp-engineNEWAPI 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. 39133
    Prev
    Next
    INDEX
    Explanations

    From

    np_max-act · gemini-2.0-flash

    The neuron detects the phrase “From the” at the start of a text segment, i.e. blocks of source‐attribution or “From the …” citations.

    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
    Decoration
    -0.08
    言う
    -0.07
     spanking
    -0.06
     ":
    -0.06
    taking
    -0.06
     AMP
    -0.06
     ':
    -0.06
    ={<
    -0.06
     MainAxisAlignment
    -0.06
     solve
    -0.06
    POSITIVE LOGITS
     From
    0.09
    From
    0.08
     from
    0.06
    Executor
    0.06
     Villa
    0.06
    яг
    0.06
    /public
    0.06
    Processed
    0.06
    бом
    0.06
    .game
    0.06
    Activations Density 0.003%

    No Known Activations