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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 6625
    Prev
    Next
    INDEX
    Explanations

    words that amplify or add intensity to a subject either in a positive or negative way

    oai_token-act-pair · gemini-2.0-flash

    variations of the word "follow"

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

    following

    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
     marié
    -0.45
    бычно
    -0.42
     dugaan
    -0.41
    isDisplayed
    -0.39
    odend
    -0.39
     spines
    -0.39
    IndentedString
    -0.39
    roke
    -0.38
    UnitTest
    -0.38
    genheiten
    -0.38
    POSITIVE LOGITS
     following
    2.03
    following
    1.91
     Following
    1.77
    Following
    1.70
     FOLLOWING
    1.63
     follow
    1.51
     follows
    1.48
    follow
    1.38
    follows
    1.36
     FOLLOW
    1.27
    Activations Density 1.643%

    No Known Activations