© 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-RES-65K
    7. 54896
    Prev
    Next
    INDEX
    Explanations

    subjective, biased, or anecdotal claims

    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_40_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
    Realistic
    0.45
    理性
    0.45
    izienz
    0.43
    瘋
    0.43
    严肃
    0.41
    <unused3>
    0.41
     lồ
    0.40
    కొ
    0.39
     సాధ
    0.39
    实在
    0.39
    POSITIVE LOGITS
     subjective
    0.98
     anecdotal
    0.91
     conjecture
    0.77
     biased
    0.77
     arbitrary
    0.70
     cherry
    0.70
     anecdotes
    0.70
     bias
    0.68
     subjekt
    0.68
     partisan
    0.67
    Activations Density 0.117%

    No Known Activations