© 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-262K
    7. 5899
    Prev
    Next
    INDEX
    Explanations

    strict adherence to rules

    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_262k_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
    u
    1.01
    d
    0.73
    на
    0.64
    daki
    0.62
    w
    0.60
     Embassy
    0.59
     etwas
    0.59
     Encode
    0.59
     بزر
    0.57
    ť
    0.57
    POSITIVE LOGITS
     strict
    1.10
     rigorous
    0.95
     stricter
    0.93
    strict
    0.92
     строго
    0.90
     STRICT
    0.90
    Strict
    0.89
     rigid
    0.86
     estrict
    0.86
     strictly
    0.85
    Activations Density 0.033%

    No Known Activations