© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 0-GEMMASCOPE-TRANSCODER-16K
    7. 10956
    Prev
    Next
    INDEX
    Explanations

    the letters 'ni' substrings in words

    oai_token-act-pair · gemini-2.0-flash

    Code, citations, non-English text

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ni
    -1.08
    mi
    -1.02
    bi
    -0.88
    si
    -0.87
    der
    -0.83
    )
    -0.81
    cas
    -0.77
     ni
    -0.75
    mo
    -0.75
    pi
    -0.73
    POSITIVE LOGITS
     myſelf
    1.77
     itſelf
    1.72
     themſelves
    1.62
     Monfieur
    1.58
    <bos>
    1.56
     Jefus
    1.50
     Efq
    1.47
     himſelf
    1.47
     houſe
    1.46
     Majefty
    1.43
    Activations Density 1.249%

    No Known Activations