© 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. 25398
    Prev
    Next
    INDEX
    Explanations

    existence and location out there

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

    The neuron is strongly activated by the word “there,” especially when it appears in existential or “out there” constructions.

    oai_token-act-pair · o4-miniTriggered 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
     and
    -2.33
    明日は
    -2.17
     pestañas
    -1.99
     this
    -1.89
    えっ
    -1.87
    もうすぐ
    -1.85
    ごめん
    -1.84
    あー
    -1.84
    をします
    -1.82
    鹀
    -1.80
    POSITIVE LOGITS
    を
    2.41
    で
    2.16
    などが
    2.08
     –
    2.05
     of
    1.85
    ですが
    1.81
    もあります
    1.78
    なので
    1.77
    と
    1.70
    もある
    1.63
    Activations Density 0.007%

    No Known Activations