© 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. 34-GEMMASCOPE-RES-131K
    4. 84369
    Prev
    Next
    INDEX
    Explanations

    happening soon

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

    The neuron fires on adverbial/time‐reference words that look ahead or promise repetition—e.g. “again,” “anytime,” “soon,” and similar markers of future recurrence.

    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
    initial
    -0.91
    previous
    -0.88
    anmoins
    -0.87
    ior
    -0.85
     ADDITIONAL
    -0.82
    好不容易
    -0.81
    aking
    -0.80
     généralement
    -0.79
    ilia
    -0.79
    lum
    -0.78
    POSITIVE LOGITS
     soon
    8.31
    Soon
    5.84
    soon
    5.72
     Soon
    5.59
     SOON
    4.47
     bientôt
    4.13
     snart
    3.64
     скоро
    3.64
    很快
    3.16
     sooner
    3.03
    Activations Density 0.150%

    No Known Activations