© 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. 40-GEMMASCOPE-2-RES-262K
    7. 4118
    Prev
    Next
    INDEX
    Explanations

    not the appropriate

    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_40_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
     принци
    0.41
     overriding
    0.39
     फं
    0.39
    動き
    0.38
     comply
    0.38
    טע
    0.37
    reetings
    0.36
    ք
    0.36
    要注意
    0.36
    Overrides
    0.36
    POSITIVE LOGITS
     appropriate
    0.84
     inappropriate
    0.84
     geeign
    0.81
     Appropriate
    0.80
    appropriate
    0.77
     vhod
    0.77
    Appropri
    0.75
    适合
    0.75
     مناسب
    0.74
     подходя
    0.71
    Activations Density 0.014%

    No Known Activations