© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 30958
    Prev
    Next
    INDEX
    Explanations

    Printer filament, polar white

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

    The neuron fires strongly on document‐header lines that name and version 3D-printer products—e.g. titles like “Marlin 3D Printer Firmware,” “Recycled 3D Printer Filament,” or “J55 3D Printer”—so it’s detecting those technical header phrases (product names plus version/date info).

    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
    icot
    -0.76
     Felsen
    -0.72
     consultato
    -0.71
     юриди
    -0.71
    刺激
    -0.69
     پارچه
    -0.68
    ๘
    -0.68
    uggets
    -0.67
    lexikon
    -0.66
    enschein
    -0.66
    POSITIVE LOGITS
     filament
    2.08
     PLA
    1.88
     Filament
    1.68
    PLA
    1.64
    fila
    1.53
    Fila
    1.41
    Cura
    1.39
     filaments
    1.36
     Cura
    1.28
     fila
    1.18
    Activations Density 0.023%

    No Known Activations