© 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. 95597
    Prev
    Next
    INDEX
    Explanations

    recurrence relations and products

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

    The neuron strongly activates on programming/operator tokens (e.g. pattern‐matching bars “|”, arrows “->”/“=>”, wildcards “_”, arithmetic operators, braces, etc.).

    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
     REPRESENTATIVES
    -0.88
     ajedrez
    -0.85
     aerolí
    -0.84
     planeta
    -0.82
    Mil
    -0.82
    میل
    -0.81
    cej
    -0.79
    nologue
    -0.79
     bonos
    -0.79
    illotine
    -0.79
    POSITIVE LOGITS
     fascic
    0.86
    itest
    0.81
     będzie
    0.79
    ificance
    0.78
    enscha
    0.77
    endor
    0.76
    destruct
    0.74
     politico
    0.73
     різних
    0.73
     scintill
    0.73
    Activations Density 0.002%

    No Known Activations