© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 10059
    Prev
    Next
    INDEX
    Explanations

    `used`, `sequence`, `label`

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

    The neuron strongly activates on verbs (often in infinitive or imperative form) that lay out steps or instructions—i.e. action words used in “how-to” guidance.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ر
    0.84
    θέ
    0.82
    mins
    0.80
     λογ
    0.77
     በመ
    0.76
    ილი
    0.76
    zust
    0.75
    ল
    0.75
     πί
    0.75
     lồ
    0.75
    POSITIVE LOGITS
    шает
    0.87
     sequence
    0.74
     inspired
    0.71
     exerts
    0.71
     sequences
    0.70
     agencies
    0.70
     fosters
    0.70
     sturdy
    0.68
     copyrighted
    0.68
     regarding
    0.66
    Activations Density 0.000%

    No Known Activations