© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 53900
    Prev
    Next
    INDEX
    Explanations

    safety guidelines, usage differences, health care

    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
    ందిన
    0.45
    രുവ
    0.44
    )])
    0.44
    ریس
    0.41
     Cmd
    0.41
    いくつか
    0.41
    㖩
    0.40
    塩素
    0.40
    つけて
    0.40
    いくつ
    0.39
    POSITIVE LOGITS
     banished
    0.46
    كَ
    0.45
     clever
    0.43
     motivator
    0.43
     quienes
    0.43
     бюро
    0.43
    kiem
    0.42
     vanqu
    0.42
     grumpy
    0.42
     человек
    0.42
    Activations Density 0.001%

    No Known Activations