© 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. 15283
    Prev
    Next
    INDEX
    Explanations

    government titles and commercial platforms

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

    The neuron is looking for numeric tokens (e.g. dates, counts, IDs—any standalone numbers).

    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
     supposedly
    -1.09
     evidently
    -1.07
    ĩnh
    -1.01
    diger
    -0.96
     usually
    -0.93
     ostensibly
    -0.93
     rencana
    -0.91
     Parlamento
    -0.90
     immensely
    -0.89
     oddly
    -0.88
    POSITIVE LOGITS
    <eos>
    1.06
    gy
    0.98
    gard
    0.92
    lie
    0.90
     CURRENT
    0.89
    ht
    0.88
    pr
    0.88
     ROCK
    0.87
    wer
    0.85
    isted
    0.85
    Activations Density 0.051%

    No Known Activations