© 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
    3. Gemma-3-4B-IT
    5. 17-GEMMASCOPE-2-RES-16K
    7. 2067
    Prev
    Next
    INDEX
    Explanations

    potential conflicts or actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-4b-it/resid_post/layer_17_width_16k_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
     тяж
    0.83
     आणि
    0.82
     আর
    0.81
     وت
    0.80
     তথা
    0.79
     και
    0.77
     nasty
    0.77
     و
    0.77
     (
    0.76
     而且
    0.76
    POSITIVE LOGITS
    0
    1.11
    5
    1.02
    4
    1.02
    2
    0.98
    1
    0.97
    3
    0.97
    6
    0.93
    8
    0.87
    7
    0.85
    9
    0.83
    Activations Density 1.227%

    No Known Activations