© 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. 22-GEMMASCOPE-RES-131K
    4. 69176
    Prev
    Next
    INDEX
    Explanations

    biological research, cyclic sweep, stress response, containing sites

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

    technical or scientific terminology in academic or specialized texts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron selectively activates on technical or scientific‐style content words (e.g. “neurons,” “biological,” “operation,” “stress,” “cyclic,” 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
    といふ
    -0.91
     obligaciones
    -0.88
    Tó
    -0.87
    っけ
    -0.84
     presque
    -0.84
     SCHOLASTIC
    -0.83
     personnalisé
    -0.83
     einander
    -0.82
     cuotas
    -0.82
    maßen
    -0.81
    POSITIVE LOGITS
    ningss
    1.02
    alisation
    1.01
    まずは
    1.01
    atization
    1.00
     ï
    0.98
     kuhusu
    0.95
    igrette
    0.94
    ningens
    0.93
    qtd
    0.91
     EXISTS
    0.90
    Activations Density 0.046%

    No Known Activations