© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 13210
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to cognition and decision-making.

    oai_token-act-pair · gemini-2.0-flash

    the word "which"

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

    which

    np_max-act-logits · gemini-2.0-flash

    which

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    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
    KindOfClass
    -0.56
     StringTokenizer
    -0.54
     MonoBehaviour
    -0.54
    ाहरण
    -0.54
    arschijnlijk
    -0.52
    fycat
    -0.51
    IVEREF
    -0.50
     تضيفلها
    -0.50
    tsy
    -0.48
     kasarigan
    -0.48
    POSITIVE LOGITS
     which
    3.83
    which
    3.22
     WHICH
    2.98
     Which
    2.97
    Which
    2.92
    hich
    1.98
     laquelle
    1.95
     lequel
    1.87
     whom
    1.77
     wich
    1.77
    Activations Density 11.973%

    No Known Activations