© 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. 22-GEMMASCOPE-RES-131K
    7. 25401
    Prev
    Next
    INDEX
    Explanations

    one day / someday / eventually

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

    The neuron flags words and phrases that indicate future time—particularly “one day”/“someday” constructions and future‐tense auxiliaries like “will” or “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_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
     horarios
    -1.25
    recently
    -1.23
     recent
    -1.23
    immediately
    -1.23
     駅前
    -1.22
     imediatamente
    -1.20
    뺏
    -1.20
     často
    -1.17
     rése
    -1.15
    recent
    -1.13
    POSITIVE LOGITS
     will
    2.11
     Eventually
    1.64
     someday
    1.48
     Maybe
    1.42
    Maybe
    1.40
    Eventually
    1.38
     irgendwann
    1.34
     maybe
    1.24
     vielleicht
    1.24
     Perhaps
    1.20
    Activations Density 0.027%

    No Known Activations