© 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. 13854
    Prev
    Next
    INDEX
    Explanations

    mystery

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the word “mystery,” especially when it appears as a heading or title.

    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
    _buf
    -0.07
    (cd
    -0.07
     beet
    -0.06
     sand
    -0.06
     incub
    -0.06
     pink
    -0.06
     rempl
    -0.06
     oct
    -0.06
    (OpCodes
    -0.06
    Sand
    -0.06
    POSITIVE LOGITS
     mystery
    0.14
     Mystery
    0.12
     mysteries
    0.11
     mystical
    0.10
     mysterious
    0.09
     Myst
    0.09
    ystery
    0.09
     myst
    0.08
     дослід
    0.08
     myster
    0.07
    Activations Density 0.009%

    No Known Activations