© 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. 15864
    Prev
    Next
    INDEX
    Explanations

    start of sequences

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

    The neuron activates on numeric tokens (digits and numbers), particularly year-like 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
     also
    -1.52
     that
    -1.34
     so
    -1.31
     will
    -1.19
     again
    -1.18
     then
    -1.17
     another
    -1.14
     on
    -1.12
     опять
    -1.02
    ではまた
    -1.02
    POSITIVE LOGITS
     ersten
    1.75
     první
    1.72
     eerste
    1.65
     пър
    1.64
     primeira
    1.64
     pertama
    1.63
     primeiro
    1.59
     första
    1.55
     first
    1.53
     commencer
    1.52
    Activations Density 0.104%

    No Known Activations