© 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. 8762
    Prev
    Next
    INDEX
    Explanations

    auxiliary verbs followed by words indicating an action

    oai_token-act-pair · gemini-2.0-flash

    variations of the word "come."

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

    come

    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
    MLLoader
    -0.58
    ثيق
    -0.57
    PreferredItem
    -0.56
    談社
    -0.55
    FieldBuilder
    -0.53
     følge
    -0.52
     Moors
    -0.52
    WebVitals
    -0.51
     RIPRODUZIONE
    -0.51
     comigo
    -0.50
    POSITIVE LOGITS
     comes
    1.02
     come
    0.98
     across
    0.94
    comes
    0.90
     Across
    0.88
     Comes
    0.85
     coming
    0.81
     ACROSS
    0.81
     COME
    0.79
     Come
    0.78
    Activations Density 0.961%

    No Known Activations