© 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. 102723
    Prev
    Next
    INDEX
    Explanations

    HTML tags

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

    The neuron activates on programming‐language syntax tokens—keywords, operators, import/register calls, angle‐bracket tags—i.e. bits of code rather than natural‐language text.

    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
     FOV
    -0.81
    মান
    -0.81
    asics
    -0.80
    noj
    -0.77
    chom
    -0.74
    彪
    -0.74
     restent
    -0.73
    reter
    -0.72
    estero
    -0.71
    rote
    -0.71
    POSITIVE LOGITS
    Zwei
    0.91
    Vorte
    0.84
    schöne
    0.84
    Fernseh
    0.83
    自家
    0.82
    了一眼
    0.82
    zwei
    0.81
    Datuak
    0.81
     app
    0.81
    können
    0.80
    Activations Density 0.040%

    No Known Activations