© 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. 34-GEMMASCOPE-RES-131K
    4. 98665
    Prev
    Next
    INDEX
    Explanations

    type, protocol, param, id, ibility

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

    This neuron fires on uppercase identifier fragments used in code (e.g. parts of macro, enum, or constant names).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     помогает
    -0.88
    jenigen
    -0.86
     Of
    -0.84
     odpowiada
    -0.82
    iots
    -0.82
     więc
    -0.81
     많은
    -0.81
    nové
    -0.80
     chiusura
    -0.79
     dringend
    -0.79
    POSITIVE LOGITS
    None
    1.32
    NONE
    1.19
    none
    1.19
     none
    1.12
     None
    1.11
     typy
    1.02
     QWidget
    1.01
     andet
    1.00
     Jezus
    0.99
     crossorigin
    0.99
    Activations Density 0.036%

    No Known Activations