© 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. 34-GEMMASCOPE-RES-131K
    7. 7259
    Prev
    Next
    INDEX
    Explanations

    reverse, backward, descending

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

    The neuron is detecting numeric literals (sequences of digits) in code or text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    FlatButton
    -0.81
    plains
    -0.71
    앙
    -0.70
     resizing
    -0.70
    ROUND
    -0.70
    ⊞
    -0.69
    árny
    -0.69
    Capítulo
    -0.68
    ἁ
    -0.68
     роки
    -0.67
    POSITIVE LOGITS
     reverse
    3.13
     backwards
    3.08
     reversed
    2.77
    reverse
    2.53
     backward
    2.50
     Reverse
    2.44
    reversed
    2.38
    Reverse
    2.34
    逆
    2.17
    [::-
    2.09
    Activations Density 0.027%

    No Known Activations