© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103737
    Prev
    Next
    INDEX
    Explanations

    future tense will/’ll

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

    The neuron detects future‐tense constructions, especially auxiliaries like “will”/“’ll” and “be going to.”

    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
     niedersachsen
    -1.98
     quadrinhos
    -1.86
    饜
    -1.83
    㩒
    -1.73
    蘩
    -1.72
     CONCRETE
    -1.69
     搞笑
    -1.68
    ⸫
    -1.66
     素描
    -1.64
    艄
    -1.63
    POSITIVE LOGITS
    as
    1.74
    </b>
    1.70
    F
    1.68
    od
    1.66
    co
    1.63
    S
    1.63
     more
    1.62
    “
    1.60
     )
    1.59
    no
    1.58
    Activations Density 0.032%

    No Known Activations