© 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. 10-GEMMASCOPE-RES-131K
    7. 99422
    Prev
    Next
    INDEX
    Explanations

    again

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

    The neuron activates on occurrences of the discourse marker “Again,” i.e. the word “Again” when it’s used to signal repetition or refer back.

    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
    {
    -1.89
    ﮕ
    -1.78
    
    -1.71
     لدي
    -1.70
    錞
    -1.68
    旄
    -1.62
    ፨
    -1.59
    扨
    -1.54
     mereka
    -1.53
    砖
    -1.53
    POSITIVE LOGITS
    _
    1.83
     kanssa
    1.61
    );
    1.48
     â
    1.45
     huom
    1.41
     showering
    1.41
    hline
    1.40
     For
    1.37
     Often
    1.37
    或者
    1.34
    Activations Density 0.003%

    No Known Activations