© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38598
    Prev
    Next
    INDEX
    Explanations

    anti-hypertension or alignment specifications

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

    The neuron activates only on the numeric token “64,” i.e. it spots the “64” literal (as in i64) in code.

    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
    その
    -1.34
     their
    -1.28
    それを
    -1.23
    all
    -1.19
    もあって
    -1.18
    そして
    -1.12
    それが
    -1.09
     åren
    -1.09
    ⋘
    -1.05
     saken
    -1.04
    POSITIVE LOGITS
    Begriffsklä
    1.30
     vimos
    1.23
     which
    1.19
    Bibliografie
    1.19
    にしました
    1.16
    leyball
    1.13
    を書く
    1.12
    kreises
    1.05
    ógica
    1.05
     nossas
    1.05
    Activations Density 0.000%

    No Known Activations