© 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. 34-GEMMASCOPE-RES-131K
    7. 2637
    Prev
    Next
    INDEX
    Explanations

    javadoc code and licenses

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

    The neuron detects three-digit numeric tokens (i.e. sequences of exactly three digits).

    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
     both
    -1.19
     to
    -0.98
     very
    -0.94
     without
    -0.87
     with
    -0.84
     whose
    -0.84
     fairly
    -0.82
    ="<?
    -0.81
    scalatest
    -0.80
     ciąży
    -0.80
    POSITIVE LOGITS
    original
    1.01
     préf
    0.95
    行きます
    0.91
    WebControls
    0.88
    ượ
    0.88
    viate
    0.87
    ifique
    0.86
     ähn
    0.85
     EARL
    0.85
    ніе
    0.85
    Activations Density 0.001%

    No Known Activations