© 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-IT
    5. 40-GEMMASCOPE-2-TRANSCODER-262K
    7. 242673
    Prev
    Next
    INDEX
    Explanations

    Identity formation and crisis

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_40_width_262k_l0_small_affine
    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
     нужен
    0.37
     PM
    0.37
     बढ़ाता
    0.36
    간다
    0.36
    uele
    0.35
    婧
    0.35
     바꿔
    0.35
    উন
    0.34
     нужна
    0.34
    atious
    0.34
    POSITIVE LOGITS
     formation
    0.48
     formed
    0.40
     Models
    0.38
     Atal
    0.38
    Formation
    0.38
     defense
    0.38
     formative
    0.38
     hacking
    0.37
    models
    0.37
     Formation
    0.37
    Activations Density 0.003%

    No Known Activations