© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 12550
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to segmenting data or objects

    oai_token-act-pair · gemini-2.0-flash

    say "into"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
     into
    -1.73
    Into
    -1.60
     Into
    -1.60
     INTO
    -1.51
    into
    -1.47
    INTO
    -1.33
     vào
    -0.83
    يديا
    -0.63
     thành
    -0.57
    nto
    -0.57
    POSITIVE LOGITS
    TagMode
    0.81
     resourceCulture
    0.73
    Xna
    0.61
     المعيارى
    0.59
     صوتيه
    0.59
    例句
    0.57
    tagext
    0.56
    Personensuche
    0.56
    gitto
    0.56
    VersionUID
    0.54
    Activations Density 1.870%

    No Known Activations