© 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. 65649
    Prev
    Next
    INDEX
    Explanations

    game items and statistics

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

    The neuron strongly activates on tokens that are numeric (i.e. containing digits, like version numbers, dates, raid times, or counts).

    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
     Weiß
    -0.82
    Belo
    -0.79
    唷
    -0.78
    ั่ว
    -0.77
    uparrow
    -0.75
    轧
    -0.75
    enski
    -0.71
     Weeds
    -0.68
     ทอง
    -0.68
    iores
    -0.68
    POSITIVE LOGITS
    ánt
    0.98
    ligente
    0.78
     amie
    0.77
    jaci
    0.76
     ?)
    0.74
     Tonga
    0.73
    VERB
    0.73
     nella
    0.72
    Vel
    0.71
     Legion
    0.70
    Activations Density 0.011%

    No Known Activations