© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 3335
    Prev
    Next
    INDEX
    Explanations

    who doesn't know

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

    The neuron fires on personal pronouns (I, you, we, anyone, etc.), i.e. words that refer to the speaker or listener.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Knew
    -1.26
    knew
    -1.23
     KNOWN
    -1.05
     Knowing
    -1.04
     Known
    -1.02
     Knows
    -1.01
    knows
    -1.00
     aware
    -1.00
     Aware
    -0.96
    knowing
    -0.94
    POSITIVE LOGITS
    很喜欢
    0.99
    坏了
    0.82
    こそ
    0.80
     героев
    0.79
     Очень
    0.76
    ȸ
    0.76
    션
    0.75
    iligen
    0.75
     triumphantly
    0.75
    either
    0.73
    Activations Density 0.049%

    No Known Activations