© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 67594
    Prev
    Next
    INDEX
    Explanations

    phrases starting with like, on, here, a

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

    The neuron activates on numeric tokens (numbers) in the text.

    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
    -1.40
     him
    -1.15
    Menurutnya
    -1.09
     his
    -1.07
    クーポン
    -1.05
     Psicología
    -1.02
     only
    -1.02
    satunya
    -1.01
     noemen
    -1.01
     Only
    -1.00
    POSITIVE LOGITS
     новый
    1.03
     getragen
    1.02
    ITY
    1.02
    клад
    1.00
     adecuados
    1.00
    émoc
    0.96
     esimerkiksi
    0.93
    蹬
    0.93
    in
    0.92
     gegenwär
    0.92
    Activations Density 0.000%

    No Known Activations