Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 21-GEMMASCOPE-TRANSCODER-16K
    4. 10963
    Prev
    Next
    INDEX
    Explanations

    the word "are" plus additional words specifying quantity

    oai_token-act-pair · gemini-2.0-flash

    quantifying terms

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

    are

    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
     the
    -0.46
     ouvriers
    -0.43
     nuages
    -0.42
     conoz
    -0.40
     présent
    -0.39
     he
    -0.38
    ifrance
    -0.38
    ggish
    -0.38
     travailleurs
    -0.37
     there
    -0.36
    POSITIVE LOGITS
     hundreds
    0.90
     dozens
    0.85
     ſeveral
    0.85
     thousands
    0.84
    MLLoader
    0.81
     kasarigan
    0.81
    hundreds
    0.81
     ligiloj
    0.80
     THOUS
    0.76
     countless
    0.74
    Activations Density 1.718%

    No Known Activations