© 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. 60520
    Prev
    Next
    INDEX
    Explanations

    cram manyspecifications nicebetween concepts

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

    avoid cramming too much

    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
    ದಲ್ಲ
    0.49
    tak
    0.47
    ợ
    0.47
    something
    0.47
     धमाल
    0.46
    ರಾ
    0.46
    நுட்ப
    0.46
    वर्ग
    0.46
    itamins
    0.46
    said
    0.45
    POSITIVE LOGITS
     dovet
    0.48
     biasing
    0.46
     between
    0.46
    Х
    0.46
     zwischen
    0.46
    вает
    0.45
     lief
    0.45
     benef
    0.45
     વચ્ચે
    0.45
     între
    0.44
    Activations Density 0.001%

    No Known Activations