© 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. 32694
    Prev
    Next
    INDEX
    Explanations

    trade-offs and sacrifices

    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
    莳
    0.38
    Truly
    0.37
    ாதை
    0.36
     rappresenta
    0.36
     అంత
    0.36
     effectively
    0.35
    ocked
    0.35
    particularly
    0.35
    randomIndex
    0.35
    idget
    0.35
    POSITIVE LOGITS
    牺牲
    0.87
     sacrificing
    0.86
     sacrifice
    0.80
     sacrifices
    0.80
    犧
    0.79
     sacrificed
    0.78
     tradeoff
    0.67
     compromising
    0.66
     बलिदान
    0.64
     compromise
    0.64
    Activations Density 0.118%

    No Known Activations