© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 1719
    Prev
    Next
    INDEX
    Explanations

    while contrast or concession

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

    The neuron is tuned to detect first-person perspective markers (e.g. “I,” “we”) and related personal-voice words.

    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
    unless
    0.77
     unless
    0.71
     если
    0.66
    если
    0.61
     hvis
    0.61
     якщо
    0.61
    if
    0.58
    .
    0.58
     nếu
    0.56
    除非
    0.56
    POSITIVE LOGITS
     ostensibly
    1.03
     superficially
    1.00
     nominally
    0.92
     technically
    0.86
     theoretically
    0.84
     seemingly
    0.75
    看似
    0.74
     somewhat
    0.70
     stále
    0.68
     undeniably
    0.68
    Activations Density 0.524%

    No Known Activations