© 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. 22-GEMMASCOPE-RES-131K
    4. 54158
    Prev
    Next
    INDEX
    Explanations

    past, present, or future statements

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

    The neuron is sensitive to words and phrases expressing expectation, intention, or prediction (e.g. modal verbs and related expressions about what will or would happen).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    currently
    -1.05
     currently
    -1.02
    Currently
    -0.93
     traductions
    -0.86
    diendo
    -0.84
    geist
    -0.82
    nisk
    -0.82
     vannak
    -0.82
    Haven
    -0.81
     controlada
    -0.81
    POSITIVE LOGITS
     this
    1.35
     would
    1.34
     was
    1.24
     ήταν
    1.13
     sẽ
    1.03
     будет
    1.02
     было
    0.98
    would
    0.98
     هذا
    0.98
     würde
    0.91
    Activations Density 0.070%

    No Known Activations