© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 8895
    Prev
    Next
    INDEX
    Explanations

    actions performed on objects

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

    The neuron detects main action words in gerund or present‐participle form (–ing verbs) used as the lead verb in sentences or headings.

    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
     industrie
    -1.43
     was
    -1.38
    –¿
    -1.30
    ”.
    -1.24
     arma
    -1.22
    '';
    -1.20
    ÉE
    -1.20
    oit
    -1.19
    ‘‘
    -1.18
     “
    -1.16
    POSITIVE LOGITS
    Following
    1.41
     such
    1.38
    ğiniz
    1.34
     your
    1.33
     it
    1.31
    Your
    1.31
     tratado
    1.30
     will
    1.27
     any
    1.27
     following
    1.22
    Activations Density 0.136%

    No Known Activations