© 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. 22-GEMMASCOPE-RES-131K
    7. 15548
    Prev
    Next
    INDEX
    Explanations

    part of a system

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

    The neuron fires strongly on scientific phrases that quantify essential or “key” components—e.g. “key component,” “essential part,” or “important element” followed by a numerical measure.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     something
    -1.20
     något
    -1.19
     time
    -1.11
     wohin
    -1.09
     frutto
    -1.08
    だそうです
    -1.07
    过来了
    -1.06
     utilizzando
    -1.05
    These
    -1.04
     silicona
    -1.04
    POSITIVE LOGITS
     of
    1.87
     in
    1.56
     our
    1.49
    li
    1.16
    其妙
    1.13
    𝟬
    1.05
     reik
    1.05
    important
    1.04
     pido
    1.03
     =
    1.01
    Activations Density 0.065%

    No Known Activations