© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 70129
    Prev
    Next
    INDEX
    Explanations

    looking forward or uncertain future

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

    The neuron is signaling tokens that occur at the start of new sentences or clauses (including the punctuation that opens them).

    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
     dziś
    -1.77
     who
    -1.59
     dzisiaj
    -1.52
     علشان
    -1.48
     dagens
    -1.43
     أثناء
    -1.38
    ADDING
    -1.38
     هنگام
    -1.38
     lorsqu
    -1.36
    簡單
    -1.29
    POSITIVE LOGITS
     it
    2.17
     this
    2.16
     for
    1.98
     to
    1.78
     will
    1.67
     its
    1.54
     there
    1.51
    子和
    1.38
    ského
    1.38
    osgi
    1.35
    Activations Density 0.143%

    No Known Activations