© 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. 22-GEMMASCOPE-RES-131K
    7. 128123
    Prev
    Next
    INDEX
    Explanations

    The neuron flags programming‐style tokens—especially code identifiers and keywords (e.g. function or class names) in source code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    sequence elgg, sequence dddf

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     Asimismo
    -1.09
    ポン
    -0.96
     cinquième
    -0.92
    阉
    -0.90
     ausein
    -0.90
     девочек
    -0.89
    ojo
    -0.87
     ewes
    -0.86
    メス
    -0.85
     hisz
    -0.84
    POSITIVE LOGITS
     from
    1.30
     another
    1.25
     after
    1.17
     without
    1.07
     when
    1.05
     like
    1.01
     to
    0.98
     because
    0.98
     difficult
    0.96
     cleverly
    0.96
    Activations Density 0.004%

    No Known Activations