© 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. 10-GEMMASCOPE-RES-131K
    7. 15099
    Prev
    Next
    INDEX
    Explanations

    to-do lists and how-tos

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

    The neuron fires on occurrences of the phrase “to-do list” (i.e. the tokens “to,” “do,” “list,” and their hyphenated variants).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ),
    -2.42
    th
    -2.03
    Following
    -2.02
    After
    -1.98
    x
    -1.95
    which
    -1.92
     allows
    -1.90
    Surprisingly
    -1.89
    Putting
    -1.87
    If
    -1.84
    POSITIVE LOGITS
    栳
    2.19
    GEBUR
    2.16
     deliciously
    1.95
    火焰
    1.90
     tohoto
    1.89
    被他
    1.87
    躥
    1.87
     luscious
    1.85
     sposob
    1.84
     椅子
    1.80
    Activations Density 0.007%

    No Known Activations