© 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. 28289
    Prev
    Next
    INDEX
    Explanations

    whether or not

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

    The neuron strongly responds to common, high‐frequency verbs (especially irregular past‐tense or gerund forms like “got,” “done,” “played,” “speaking,” “opens,” and “times”).

    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
    Angebot
    -1.82
     शुर
    -1.63
     furthering
    -1.62
     nestled
    -1.55
     bicicleta
    -1.52
     blatantly
    -1.51
     because
    -1.51
    США
    -1.50
     from
    -1.49
     Providing
    -1.48
    POSITIVE LOGITS
    '
    2.03
     relentless
    1.60
    ,
    1.55
     joyful
    1.54
    _
    1.54
     strikingly
    1.52
     astonishing
    1.51
     astounding
    1.50
     joyous
    1.49
     overly
    1.48
    Activations Density 0.172%

    No Known Activations