© 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. 41-GEMMASCOPE-2-RES-262K
    7. 32927
    Prev
    Next
    INDEX
    Explanations

    mode collapse

    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_all/layer_41_width_262k_l0_big
    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
    wage
    0.59
    IENCE
    0.55
    ারি
    0.55
    ো
    0.52
    ENGE
    0.51
    domain
    0.50
    ಂತೆ
    0.50
    urine
    0.47
    ethod
    0.47
    ิค
    0.47
    POSITIVE LOGITS
    에서
    0.45
     for
    0.42
    切换
    0.42
    ll
    0.41
    חת
    0.41
     Mode
    0.41
    (
    0.41
    ]
    0.41
     mode
    0.40
    할
    0.40
    Activations Density 0.002%

    No Known Activations