© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 59162
    Prev
    Next
    INDEX
    Explanations

    recommendations

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

    The neuron fires for multi‐digit number tokens—e.g. citation years, page or article numbers, grant or docket identifiers—i.e. long numeric strings.

    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
     will
    -1.10
     in
    -1.04
     all
    -1.03
     so
    -1.02
     focusing
    -1.00
     had
    -1.00
     may
    -0.94
     says
    -0.94
     also
    -0.93
     ALL
    -0.93
    POSITIVE LOGITS
    recommend
    1.09
     entusias
    1.05
     persegu
    0.98
    קות
    0.96
    その他
    0.96
     quatrième
    0.93
    atique
    0.92
     exuber
    0.91
     fiasco
    0.91
     vinte
    0.91
    Activations Density 0.003%

    No Known Activations