© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 1815
    Prev
    Next
    INDEX
    Explanations

    grammatical articles and connecting prepositions

    oai_token-act-pair · gemini-2.0-flash

    references to the word 'garage'

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

    various languages

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    boldmath
    -0.55
    numerusform
    -0.54
     preferências
    -0.54
    EDEFAULT
    -0.52
    ikan
    -0.52
    UIControlState
    -0.51
    dingen
    -0.51
    NESOTA
    -0.51
    Datuak
    -0.50
     Mehmet
    -0.48
    POSITIVE LOGITS
    garage
    0.57
     garage
    0.57
     שוליים
    0.54
     Garage
    0.54
    Garage
    0.54
    TargetException
    0.52
    market
    0.51
    oa̍t
    0.50
     SafeMath
    0.50
     market
    0.50
    Activations Density 0.400%

    No Known Activations