© 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. 15060
    Prev
    Next
    INDEX
    Explanations

    single letters, numbers, or foreign words

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

    The neuron strongly reacts to isolated single-character tokens—especially uppercase letters or digits (e.g. “A” or “1”).

    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
     científico
    -1.42
     desmotiv
    -1.17
    malte
    -1.13
    artik
    -1.12
    lampu
    -1.06
    ciri
    -1.05
     psicológico
    -1.02
     frág
    -1.02
     crítico
    -0.99
    mesin
    -0.99
    POSITIVE LOGITS
    当我
    1.08
     서비스
    1.05
     teneb
    1.03
    lité
    1.03
     VOY
    1.02
     séances
    1.01
     chargés
    1.01
     fromage
    1.01
     lati
    1.01
    ěj
    1.00
    Activations Density 0.000%

    No Known Activations