© 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. 48172
    Prev
    Next
    INDEX
    Explanations

    before, prior

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

    The neuron strongly fires on words that occur in news‐style photo captions framing violent events—especially the temporal phrase “moments before” (and its surrounding nouns like “man” or “hand”).

    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
     BEFORE
    -1.56
    Before
    -1.55
    before
    -1.49
    AFTER
    -1.41
    BEFORE
    -1.31
    After
    -1.29
    Prior
    -1.27
     PRIOR
    -1.25
    prior
    -1.23
     Before
    -1.23
    POSITIVE LOGITS
     ผ
    0.93
    ruling
    0.93
     ทำ
    0.90
    fible
    0.89
    ὠ
    0.88
    ichier
    0.87
    Anita
    0.84
    这也
    0.83
     també
    0.83
    toll
    0.83
    Activations Density 0.034%

    No Known Activations