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. 21-GEMMASCOPE-TRANSCODER-16K
    4. 5996
    Prev
    Next
    INDEX
    Explanations

    words where there is a vowel immediately after the letter 'l', 'w', 's', 'g','r', 't', 'p', or 'c', that are at least three letters long

    oai_token-act-pair · gemini-2.0-flash

    words with 'bl' or 'pl', particularly names or descriptors

    np_token-act-pair-logits · gpt-4o-mini

    bl

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope/layer_21/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
     defaultstate
    -0.90
     initComponents
    -0.89
    ंदीखरीदारी
    -0.86
    SourceChecksum
    -0.84
     okuyayım
    -0.80
    OGND
    -0.77
    ⓘ
    -0.75
     resourceCulture
    -0.72
     ویکی‌پدی
    -0.71
     betweenstory
    -0.71
    POSITIVE LOGITS
    端
    0.41
    ~
    0.40
    OutOfBounds
    0.39
    igin
    0.38
     Over
    0.38
     Fred
    0.37
     over
    0.37
     somma
    0.37
     Hor
    0.36
     t
    0.35
    Activations Density 0.604%

    No Known Activations