© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 25767
    Prev
    Next
    INDEX
    Explanations

    learning and understanding concepts

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

    The neuron strongly activates on present‐participle verb forms ending in “-ing,” i.e. gerunds that denote ongoing actions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    gerunds and present participles (words ending in "-ing") that describe learning, understanding, or mental processes.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    language about acquiring knowledge or insight and other cognitive processes, often in educational or research contexts.

    oai_token-act-pair · gpt-5Triggered 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
     dueños
    -0.87
    直播间
    -0.84
     دیکھیے
    -0.82
     семье
    -0.80
     consideran
    -0.80
     Eltern
    -0.80
    contrar
    -0.79
    immediately
    -0.79
    boné
    -0.79
     huella
    -0.78
    POSITIVE LOGITS
     ourselves
    1.04
     from
    1.02
     himself
    0.96
     самому
    0.83
     when
    0.83
     selber
    0.82
    Premiere
    0.82
    ERSON
    0.80
    RAK
    0.77
     üçün
    0.76
    Activations Density 0.077%

    No Known Activations