© 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. 4806
    Prev
    Next
    INDEX
    Explanations

    ordinal adjectives preceding superlatives or comparatives like "one of the best"

    oai_token-act-pair · gemini-2.0-flash

    comparative terms related to quality

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

    the

    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
    AndEndTag
    -0.57
    homonymie
    -0.52
    CompleteListener
    -0.52
     nød
    -0.51
    存于互联网档案馆
    -0.50
     travailleurs
    -0.49
     consommateurs
    -0.49
    ıkları
    -0.46
    ův
    -0.45
     spectateurs
    -0.45
    POSITIVE LOGITS
     many
    0.94
     most
    0.81
    tagHelper
    0.78
     more
    0.77
     few
    0.75
     متعلقه
    0.71
     best
    0.71
     better
    0.69
    CodeAttribute
    0.68
    LabelTagHelper
    0.66
    Activations Density 2.551%

    No Known Activations