© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 104663
    Prev
    Next
    INDEX
    Explanations

    new in code

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

    The neuron is detecting occurrences of the “new” keyword in code, i.e. object‐instantiation expressions.

    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
    edere
    -1.20
    分かります
    -1.20
     disposição
    -1.16
    ícias
    -1.12
    揪
    -1.11
    carded
    -1.08
    ALOG
    -1.06
    说过
    -1.05
    わかります
    -1.05
    ļi
    -1.03
    POSITIVE LOGITS
    -
    1.12
     that
    1.00
     such
    0.98
     this
    0.95
     with
    0.95
    太大
    0.90
     dår
    0.89
     so
    0.89
     battled
    0.88
    twór
    0.86
    Activations Density 0.005%

    No Known Activations