© 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. 22-GEMMASCOPE-RES-131K
    7. 67217
    Prev
    Next
    INDEX
    Explanations

    control states

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

    The neuron activates on numeric literals—integers, decimals, percentages, or other standalone number tokens.

    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
     awd
    -0.85
     lincoln
    -0.84
     autorità
    -0.83
     avenger
    -0.83
     esigenze
    -0.83
     milano
    -0.82
     indonesia
    -0.82
     loon
    -0.81
     egret
    -0.81
     rq
    -0.80
    POSITIVE LOGITS
     null
    1.07
     fichas
    0.95
     indifer
    0.93
     downright
    0.88
     достижения
    0.85
     mixtures
    0.84
     particuli
    0.82
     importantly
    0.82
     jovem
    0.81
     typelib
    0.81
    Activations Density 0.028%

    No Known Activations