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. 5971
    Prev
    Next
    INDEX
    Explanations

    non-English words or phrases with a Southeast Asian origin, possibly Malay

    oai_token-act-pair · gemini-2.0-flash

    prefixes indicating written language or slang

    np_token-act-pair-logits · gpt-4o-mini
    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
    ⓧ
    -0.68
     purpoſe
    -0.57
    kuuta
    -0.56
     nhật
    -0.54
    bewerken
    -0.53
     houſe
    -0.53
     fubject
    -0.52
     visibilité
    -0.52
    Referencies
    -0.51
    gameserver
    -0.50
    POSITIVE LOGITS
    localctx
    0.66
     gom
    0.64
     dem
    0.64
     wet
    0.61
     na
    0.61
     no
    0.61
    Wet
    0.60
     sey
    0.59
     Say
    0.59
     se
    0.57
    Activations Density 0.218%

    No Known Activations