© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 23906
    Prev
    Next
    INDEX
    Explanations

    image compression formats

    np_acts-logits-general · gemini-2.5-flash-lite

    data compression and quality

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    撩
    0.51
    rightbar
    0.49
    гел
    0.47
     दिलाया
    0.44
    🏫
    0.44
    })^{
    0.44
    सिला
    0.44
    갑
    0.44
    哞
    0.43
     seniority
    0.43
    POSITIVE LOGITS
     compression
    1.22
     Compression
    1.17
     JPEG
    1.12
    压缩
    1.09
    Compression
    1.09
     compress
    1.08
     compressing
    1.05
     lossless
    1.03
    compression
    1.02
    Compress
    1.02
    Activations Density 0.095%

    No Known Activations