© 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-65K
    7. 63877
    Prev
    Next
    INDEX
    Explanations

    howThe phrase "how" appears frequently in `TOKENS_AFTER_MAX_ACTIVATING_TOKEN` and is also present in `TOP_ACTIVATING_TEXTS`. It also appears as a token in `MAX_ACTIVATING_TOKENS`. This seems to be the strongest recurring element. how

    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_65k_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
     and
    0.49
     ribs
    0.46
    bon
    0.45
    >>&
    0.44
     beet
    0.43
     music
    0.43
     hugged
    0.42
     saute
    0.42
     bur
    0.42
     hash
    0.42
    POSITIVE LOGITS
    𒀝
    0.55
     эффектив
    0.54
    abul
    0.52
     प्रभावी
    0.52
     законода
    0.51
    ttino
    0.51
    норийска
    0.49
    abilă
    0.48
     активность
    0.48
     партнер
    0.48
    Activations Density 0.001%

    No Known Activations