© 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. 3017
    Prev
    Next
    INDEX
    Explanations

    phrases indicating consistency of behavior or state

    oai_token-act-pair · gemini-2.0-flash

    frequent use of "always"

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

    always

    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
     always
    -0.86
     all
    -0.85
    always
    -0.79
     usually
    -0.77
     wszyscy
    -0.76
    ALWAYS
    -0.75
     tutte
    -0.74
     everyone
    -0.71
     ALWAYS
    -0.71
     every
    -0.70
    POSITIVE LOGITS
    enumi
    0.66
    Diweddarwch
    0.58
    paar
    0.56
    Vaya
    0.56
    TestTools
    0.53
    ніципалі
    0.52
     <=",
    0.51
    RenderAtEndOf
    0.51
     Looks
    0.50
     looks
    0.49
    Activations Density 1.736%

    No Known Activations