© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 110902
    Prev
    Next
    INDEX
    Explanations

    HDF, H5, Realm, array, <?php

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

    The neuron fires on shorthand technical identifiers—namely uppercase acronyms or mixed alphanumeric tokens (e.g. “.h5,” “HDF,” class names, 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_10/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
    ara
    -1.69
    ens
    -1.66
    S
    -1.55
     internazionali
    -1.55
     if
    -1.51
    含ま
    -1.50
     制作
    -1.47
    schon
    -1.46
    ons
    -1.45
    Emails
    -1.45
    POSITIVE LOGITS
     verdaderos
    1.63
    晴れ
    1.49
     vastly
    1.44
     znacznie
    1.44
     zwykle
    1.42
     nurture
    1.42
    癀
    1.41
     zupełnie
    1.41
    })=
    1.39
    婚礼
    1.38
    Activations Density 0.005%

    No Known Activations