© 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. 10-GEMMASCOPE-RES-131K
    7. 70222
    Prev
    Next
    INDEX
    Explanations

    code and technical terms

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

    This neuron is strongly activated by “=” when it’s assigning a numeric literal (i.e. numeric parameter assignments).

    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
     vedno
    -1.57
     nekaj
    -1.55
    ‘
    -1.52
     “
    -1.52
    也有
    -1.48
    H
    -1.44
     sometimes
    -1.41
    ang
    -1.40
     vendar
    -1.39
    POSITIVE LOGITS
     tóxic
    1.91
     informaron
    1.83
    "",
    1.71
    ᯾
    1.69
     timmar
    1.63
    formatie
    1.63
    缑
    1.61
    佺
    1.60
    茧
    1.57
    各有
    1.53
    Activations Density 0.014%

    No Known Activations