Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-3-27B-IT
    3. 40-GEMMASCOPE-2-RES-262K
    4. 20502
    Prev
    Next
    INDEX
    Explanations

    "the" followed by specific nouns

    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_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
     наличие
    0.43
     наличи
    0.42
     inability
    0.41
     ঘিরে
    0.41
     more
    0.41
     usage
    0.40
     உதவும்
    0.40
     revealed
    0.39
    理由
    0.39
     use
    0.38
    POSITIVE LOGITS
     ಯಾರ
    0.44
     doors
    0.43
     entsprechende
    0.43
     일이
    0.43
     ಏನ
    0.42
    的事情
    0.41
     puertas
    0.40
     abz
    0.40
     empezamos
    0.40
    を開始
    0.40
    Activations Density 0.028%

    No Known Activations