© 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. 10-GEMMASCOPE-RES-131K
    7. 90663
    Prev
    Next
    INDEX
    Explanations

    iteration and iterators

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

    The neuron strongly activates on any token containing the sequence “iter,” e.g. words like iterator, iterate, iterable, iteration, etc.

    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
    -2.91
    ’
    -2.55
    ⡦
    -2.48
    carteles
    -2.39
    actividades
    -2.36
    selbe
    -2.33
    “
    -2.31
     tetapi
    -2.27
     químicas
    -2.25
    鸯
    -2.23
    POSITIVE LOGITS
    [
    3.00
    ]
    2.69
    ный
    2.67
    {
    2.67
          
    2.63
    ),
    2.56
    ...
    2.41
                    
    2.34
     của
    2.34
    ский
    2.31
    Activations Density 0.005%

    No Known Activations