© 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. 23683
    Prev
    Next
    INDEX
    Explanations

    legal citations and license URLs

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

    The neuron activates on numeric tokens (e.g. version numbers, counts, “6,” “12,” “282,” 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_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
    唷
    -0.65
     Độ
    -0.64
    aternary
    -0.64
     Schritt
    -0.63
     pleased
    -0.63
     mũi
    -0.63
     comprehended
    -0.62
     włas
    -0.62
    ogen
    -0.61
    umpy
    -0.61
    POSITIVE LOGITS
    ndal
    0.74
    Charged
    0.73
     kapas
    0.72
     elektr
    0.71
     conductors
    0.70
     collision
    0.70
     martial
    0.69
     sanitation
    0.69
    早就
    0.69
    zyka
    0.68
    Activations Density 0.076%

    No Known Activations