© 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. 22-GEMMASCOPE-RES-131K
    7. 74929
    Prev
    Next
    INDEX
    Explanations

    music artist

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

    The neuron strongly activates on standalone technical identifiers—i.e. tokens that look like code or system names (variable names, camelCase or ALL-CAPS labels, and other programming-style identifiers).

    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
     pyta
    -1.13
     Anreg
    -1.05
     Bedürfnisse
    -1.02
     sichere
    -1.01
     zask
    -0.99
     couv
    -0.98
     schnelle
    -0.98
     rê
    -0.97
     awesome
    -0.97
     téléph
    -0.96
    POSITIVE LOGITS
     végétale
    1.17
    ขอบคุณ
    1.16
    鑿
    1.11
    NCIA
    1.09
     étanche
    1.08
    Ừ
    1.04
    ----</
    1.03
     predomin
    1.03
    itzar
    1.02
     Longitude
    1.01
    Activations Density 0.001%

    No Known Activations