© 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. 10-GEMMASCOPE-RES-131K
    7. 42796
    Prev
    Next
    INDEX
    Explanations

    numbers starting with 6 or 86

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

    The neuron strongly activates on the specific numeric token “86,” i.e. it detects occurrences of the two‐digit sequence 86.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     使っ
    -2.34
    ↵↵
    -2.16
    ).
    -2.08
     "
    -2.06
    borderSide
    -2.02
    '
    -2.02
     includes
    -1.87
     ("
    -1.79
     induces
    -1.79
    鼕
    -1.77
    POSITIVE LOGITS
     artikke
    2.13
    媄
    2.03
    翃
    2.02
     Former
    2.00
     potrze
    1.86
    が必要です
    1.84
     siitä
    1.84
     erila
    1.84
    Mostly
    1.84
     About
    1.84
    Activations Density 0.016%

    No Known Activations