© 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. 98434
    Prev
    Next
    INDEX
    Explanations

    verbs followed by direction

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

    The neuron is primarily activated by action verbs.

    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
     although
    -1.23
     only
    -1.20
     possibilité
    -1.19
    そのため
    -1.18
     help
    -1.14
    回来了
    -1.14
     also
    -1.12
     other
    -1.11
     others
    -1.10
    Чтобы
    -1.09
    POSITIVE LOGITS
     and
    2.86
    kopi
    1.19
    produzione
    1.18
     gestorben
    1.16
    ZIONE
    1.07
    kanan
    1.05
    lepaskan
    1.03
    たる
    1.00
     aanbied
    0.98
     eenige
    0.97
    Activations Density 0.038%

    No Known Activations