© 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. 9718
    Prev
    Next
    INDEX
    Explanations

    lower lip, North America, beta acid, negative reversals, light orange, destructive interference, Female Athletes

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

    respectively, separately, difference, each

    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
    ʋ
    0.43
     lebt
    0.41
     काळ
    0.40
     அதே
    0.39
    echolog
    0.38
     गेला
    0.38
     وين
    0.38
    類似
    0.38
    9
    0.38
    ist
    0.38
    POSITIVE LOGITS
     sides
    0.52
     respectively
    0.51
    それぞれ
    0.49
     각각
    0.48
    分别
    0.47
     respectivamente
    0.44
    的区别
    0.44
    respectively
    0.43
     hemispheres
    0.42
    各有
    0.42
    Activations Density 0.027%

    No Known Activations