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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 20-GEMMASCOPE-TRANSCODER-16K
    4. 14077
    Prev
    Next
    INDEX
    Explanations

    first or third person pronouns followed by verbs that indicate wanting, needing or having an opinion of something

    oai_token-act-pair · gemini-2.0-flash

    Personal experiences

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope/layer_20/width_16k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    спу
    -0.46
     few
    -0.46
    ogo
    -0.45
     مشين
    -0.43
    Clamp
    -0.42
    스터
    -0.42
     Title
    -0.41
    atorics
    -0.41
     felici
    -0.41
    vale
    -0.40
    POSITIVE LOGITS
     Мексичка
    0.71
    mybatisplus
    0.69
    ukone
    0.68
    orsese
    0.65
    InjectAttribute
    0.64
    writeField
    0.60
     IndexPath
    0.59
    Superview
    0.57
    copg
    0.57
     mukana
    0.56
    Activations Density 1.424%

    No Known Activations