© 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. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 899
    Prev
    Next
    INDEX
    Explanations

    focus on

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

    The neuron flags unusually common (“high‐frequency”) tokens, i.e. words that recur often in a given text.

    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
    торой
    -0.90
     utilise
    -0.87
    ologio
    -0.85
    月から
    -0.85
    krivning
    -0.83
    +};
    -0.81
    NewLabel
    -0.81
    Tá
    -0.80
     *\
    -0.80
    Bibliograf
    -0.79
    POSITIVE LOGITS
     focus
    7.06
    focus
    5.91
     focuses
    5.50
     focusing
    5.34
    Focus
    5.34
     focused
    5.13
     Focus
    5.00
     concentrate
    4.56
    focused
    4.41
     Focusing
    4.41
    Activations Density 0.244%

    No Known Activations