© 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. 10-GEMMASCOPE-RES-131K
    7. 45335
    Prev
    Next
    INDEX
    Explanations

    midway through

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

    The neuron fires strongly on occurrences of the word “through” (and closely related forms like “halfway through”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     possible
    -2.08
     more
    -1.98
     has
    -1.98
     slight
    -1.95
     tomado
    -1.95
     weitgehend
    -1.91
    ;
    -1.80
    ↵↵
    -1.76
     größeren
    -1.75
     stupid
    -1.74
    POSITIVE LOGITS
     durer
    2.19
    潆
    2.13
     salari
    2.08
    푣
    2.08
     roul
    2.02
     dieß
    1.98
    鹛
    1.98
     carrefour
    1.91
    욥
    1.91
     hiver
    1.88
    Activations Density 0.004%

    No Known Activations