© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 3247
    Prev
    Next
    INDEX
    Explanations

    bead charms

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

    The neuron fires strongly on proper names and product/brand terms in an advertising‐style listing—especially jewelry and accessory names.

    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
     molybdenum
    -0.76
    աստ
    -0.72
    ktur
    -0.72
    hoben
    -0.71
     adrenaline
    -0.68
     brome
    -0.68
     Mih
    -0.68
     logarithms
    -0.67
    INR
    -0.67
     nucleotides
    -0.67
    POSITIVE LOGITS
     beads
    2.56
     rosary
    2.14
     Beads
    2.06
     Rosary
    2.05
    beads
    1.97
     bead
    1.89
    Bead
    1.80
     ros
    1.70
    bead
    1.66
     Bead
    1.55
    Activations Density 0.013%

    No Known Activations