© 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. 111476
    Prev
    Next
    INDEX
    Explanations

    conditional exit

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

    line breaks or statement terminators immediately before control flow returns or loops continue in code.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron detects indentation levels (leading spaces) in code lines.

    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
     of
    -1.60
    \_
    -1.56
    她和
    -1.45
     ("
    -1.32
     \
    -1.32
    iempos
    -1.23
    Posté
    -1.20
     atypical
    -1.16
     else
    -1.14
     not
    -1.13
    POSITIVE LOGITS
    </i>
    1.58
    )」
    1.45
     Würde
    1.45
     muerta
    1.45
    dő
    1.44
                               
    1.41
    These
    1.37
     сделали
    1.35
    RICAL
    1.35
                         
    1.35
    Activations Density 0.013%

    No Known Activations