© 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. 65203
    Prev
    Next
    INDEX
    Explanations

    way through

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

    The neuron reliably fires on uses of “way” (often in collocations like “your way through,” “burn your way through,” “sniff their way through,” “purl your way to,” etc.) that express progression, traversal, or movement through something.

    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
     is
    -1.04
     you
    -1.02
     should
    -0.98
    目で
    -0.97
    ในการ
    -0.95
    P
    -0.93
     has
    -0.93
     used
    -0.92
    快速的
    -0.90
     can
    -0.90
    POSITIVE LOGITS
     into
    1.53
    一款
    1.21
     oblivion
    1.19
     În
    1.18
     brillantes
    1.16
     telefonu
    1.16
     دقیقه
    1.13
     vitales
    1.13
     bekämp
    1.12
     جمعه
    1.06
    Activations Density 0.082%

    No Known Activations