© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 97934
    Prev
    Next
    INDEX
    Explanations

    as published

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

    This neuron responds to the standalone words “as” (and often paired with “is”) in legal or software‐license boilerplate (e.g. the “AS IS” disclaimer).

    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
     from
    -0.93
     by
    -0.90
     función
    -0.87
     only
    -0.87
     at
    -0.84
    タ
    -0.83
     on
    -0.83
    (゚
    -0.82
     prevents
    -0.81
     even
    -0.81
    POSITIVE LOGITS
     WER
    0.98
    ΄
    0.97
     ∞
    0.97
    ργ
    0.96
    ilius
    0.96
     ‎
    0.94
     nerd
    0.94
     toutefois
    0.93
     DURING
    0.91
    uristic
    0.90
    Activations Density 0.001%

    No Known Activations