© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 2313
    Prev
    Next
    INDEX
    Explanations

    uses of the word "fake" and words associated with legal sales or transfers of ownership

    oai_token-act-pair · gemini-2.0-flash

    variations of the word "does"

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

    does/doesn't

    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
    Not in Any Lists

    No Comments

    Negative Logits
     Waray
    -0.67
     تانيه
    -0.58
    脚注の使い方
    -0.56
    曖昧さ回避
    -0.56
     विश्वसनीयता
    -0.54
    cydow
    -0.50
     privés
    -0.50
    Vidite
    -0.50
    irements
    -0.49
    Atsauces
    -0.49
    POSITIVE LOGITS
     doesn
    1.65
     does
    1.39
     Doesn
    1.38
    doesn
    1.38
     don
    1.36
    Does
    1.29
    Doesn
    1.29
     Does
    1.28
    does
    1.27
     doesnt
    1.26
    Activations Density 1.301%

    No Known Activations