© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 25464
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions or cryptographic keys

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

    This neuron responds to single-character non‐numeric tokens (e.g. one‐letter variables and symbols like “a”, “r”, “t”, “+”, “/”, “{”).

    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
     and
    -1.02
     with
    -0.99
     것입니다
    -0.96
     on
    -0.93
     решил
    -0.90
     exalt
    -0.89
    ต่างๆ
    -0.89
    ที่ดี
    -0.89
     comunale
    -0.89
    ตัวเอง
    -0.87
    POSITIVE LOGITS
     maî
    1.13
    쐈
    1.05
     jaunes
    1.05
     два
    1.02
    ̴
    1.02
    镲
    1.02
    锘
    1.01
     cuire
    1.01
    ʏ
    1.00
     téléphone
    0.98
    Activations Density 0.000%

    No Known Activations