© 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. 65973
    Prev
    Next
    INDEX
    Explanations

    optimal solutions and bounds

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

    The neuron activates on numeric and formulaic tokens (numbers, mathematical symbols, and related function words) in technical/math‐heavy contexts.

    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
    χύ
    -0.90
    ziała
    -0.86
    thong
    -0.80
    nappropriate
    -0.79
     PRED
    -0.77
    ggy
    -0.76
     temática
    -0.74
    iParam
    -0.74
    ěli
    -0.73
    estu
    -0.73
    POSITIVE LOGITS
     optimal
    3.53
     optimum
    3.06
     Optimal
    2.64
    Optimal
    2.64
    optimal
    2.64
     оптима
    2.58
     Optimum
    2.14
     óp
    2.11
     solution
    2.02
     optimale
    2.02
    Activations Density 0.032%

    No Known Activations