© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-65K
    7. 8018
    Prev
    Next
    INDEX
    Explanations

    Sometimes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_65k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     constantly
    0.54
     постоянно
    0.54
    i
    0.54
    ópez
    0.52
    vergne
    0.50
    O
    0.47
     waren
    0.47
    篭
    0.46
     probablement
    0.45
     था
    0.45
    POSITIVE LOGITS
     бывает
    0.95
     Sometimes
    0.77
    Sometimes
    0.74
     sometimes
    0.72
     Иногда
    0.71
    sometimes
    0.71
    どうしても
    0.70
     manchmal
    0.69
     бывают
    0.68
    м
    0.66
    Activations Density 0.059%

    No Known Activations