© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 38823
    Prev
    Next
    INDEX
    Explanations

    hidden instructions and unnecessary repetition

    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_262k_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
     tinc
    0.51
    栦
    0.51
    }}=
    0.49
    幵
    0.48
    প
    0.46
    or
    0.46
    僟
    0.45
    
    0.45
    ilte
    0.44
    भ
    0.43
    POSITIVE LOGITS
     strengths
    0.50
     quintessential
    0.49
     pivotal
    0.45
     percol
    0.45
    ,
    0.44
     capabilities
    0.43
     competencies
    0.43
     мощность
    0.43
     permeates
    0.43
     инфраструк
    0.42
    Activations Density 0.001%

    No Known Activations