© 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. 69282
    Prev
    Next
    INDEX
    Explanations

    absence or specific nouns

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

    The neuron fires strongly on programming-style tokens—e.g. function names, command-line keywords, code fragments, and other technical code 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
    licants
    -0.89
    riac
    -0.85
     spesso
    -0.84
    ixa
    -0.82
    ercizi
    -0.79
     vacaciones
    -0.79
    krivelse
    -0.79
     sempre
    -0.79
    siano
    -0.79
     úrov
    -0.79
    POSITIVE LOGITS
     without
    0.98
    Without
    0.88
     anyone
    0.82
    過ごす
    0.82
    without
    0.78
     Without
    0.76
     CSI
    0.76
    scrollTop
    0.75
    outines
    0.75
     anybody
    0.75
    Activations Density 0.120%

    No Known Activations