© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 1501
    Prev
    Next
    INDEX
    Explanations

    title and subtitle labels

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

    The neuron primarily activates on numeric tokens—i.e. strings containing digits (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_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
     Роб
    -0.86
    ugget
    -0.83
     departs
    -0.81
    ༄
    -0.78
    𝐲
    -0.77
     Graph
    -0.77
    🛡
    -0.75
     κι
    -0.74
    Dough
    -0.73
    ooks
    -0.73
    POSITIVE LOGITS
     subtitle
    0.90
    ▮
    0.88
    Remix
    0.84
    đu
    0.79
     obviously
    0.79
    ℏ
    0.77
     jelas
    0.77
     doubtless
    0.77
     Subtitle
    0.77
     whenever
    0.74
    Activations Density 0.035%

    No Known Activations