© 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. 22-GEMMASCOPE-RES-131K
    4. 84436
    Prev
    Next
    INDEX
    Explanations

    date and time formats

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

    The neuron fires on tokens involved in formatting functions or format‐string specifiers (e.g. calls to `.format`, date format patterns, `to_char`, `SimpleDateFormat`, 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_22/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
     alcune
    -1.34
    这种
    -1.23
     với
    -1.18
     anuncia
    -1.16
     denuncias
    -1.14
     quando
    -1.11
     alcuni
    -1.11
    這種
    -1.09
     acoper
    -1.09
     possibilità
    -1.08
    POSITIVE LOGITS
     “
    1.25
     ''
    1.23
     ”
    1.10
     septembre
    1.05
     τὴν
    1.05
    tigs
    1.05
    番茄
    1.02
     (
    1.02
     exquisitely
    1.02
     HUGE
    1.01
    Activations Density 0.005%

    No Known Activations