© 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. 2863
    Prev
    Next
    INDEX
    Explanations

    the phrase "can be" or "could be", possibly with an extra word such as "also" or "and"

    oai_token-act-pair · gemini-2.0-flash

    terms related to potential or possibility

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

    be

    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
     can
    -1.47
    can
    -1.22
    Can
    -1.18
     Can
    -1.11
     dapat
    -0.93
     können
    -0.91
     bisa
    -0.87
     CAN
    -0.84
     можно
    -0.84
    可以
    -0.83
    POSITIVE LOGITS
     føl
    0.59
    Clik
    0.53
     eccl
    0.51
     découver
    0.50
     renforcé
    0.49
     stället
    0.49
     remarqué
    0.47
     constaté
    0.47
     fédé
    0.47
     rencontr
    0.47
    Activations Density 0.480%

    No Known Activations