© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 24953
    Prev
    Next
    INDEX
    Explanations

    parts of mathematical expressions

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

    The neuron fires on occurrences of the copula “is” (and similar linking words like “that” or “if”), flagging simple present-tense linking or conditional markers.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    uyler
    -0.86
    ívar
    -0.86
    注
    -0.86
    פע
    -0.84
    putar
    -0.81
     Acesso
    -0.80
    想必
    -0.80
    <0x98>
    -0.80
    الش
    -0.80
    .__
    -0.79
    POSITIVE LOGITS
    thod
    0.94
     voeten
    0.90
     doigts
    0.89
     przew
    0.88
    ־ה
    0.84
    ếc
    0.84
     look
    0.81
     pacientes
    0.81
    obacz
    0.80
    mmers
    0.79
    Activations Density 0.009%

    No Known Activations