© 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. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 117352
    Prev
    Next
    INDEX
    Explanations

    words starting with Hark, Harc, Harg, or Herak

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on tokens beginning with “Har” (e.g. Harkers, Harkness, Hargis, Haruka).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     salat
    -0.85
    abilidad
    -0.75
    Режи
    -0.75
    ilians
    -0.73
     tev
    -0.71
     koo
    -0.71
    úmenes
    -0.71
    retVal
    -0.71
     Cowell
    -0.70
    Gum
    -0.70
    POSITIVE LOGITS
    зву
    0.73
    /"
    0.66
    tead
    0.66
     пород
    0.66
    ss
    0.65
     đi
    0.62
    ymbol
    0.62
     đầy
    0.61
     Rayleigh
    0.60
     oxidative
    0.60
    Activations Density 0.045%

    No Known Activations