© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 111167
    Prev
    Next
    INDEX
    Explanations

    conditional programming keywords

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

    method calls or function names that check for existence or boolean conditions (like `exists()`, `is_array()`, `is_object()`, `contains()`, `matches()`, `is()`, `hasValue()`).

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

    The neuron flags calls to predicate‐style functions or methods whose names begin with “is” (e.g. is_array, isdigit, exists, in_array).

    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
    na
    -1.78
     but
    -1.74
     “
    -1.67
     for
    -1.66
    us
    -1.56
     when
    -1.52
    li
    -1.45
    n
    -1.44
     and
    -1.42
    ia
    -1.32
    POSITIVE LOGITS
     görüntüsü
    1.82
     there
    1.64
     Autó
    1.51
     spørsmål
    1.51
    agramm
    1.50
     giyim
    1.48
     hjelpe
    1.48
     mieszkań
    1.45
     Offisielt
    1.43
    imbangan
    1.43
    Activations Density 0.012%

    No Known Activations