© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 3802
    Prev
    Next
    INDEX
    Explanations

    question words, especially "how" and "what", but with some common words like 'is', 'that', 'it', 'to', 'be', etc. thrown in.

    oai_token-act-pair · gemini-2.0-flash

    scenarios and people

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
     bagaimana
    -0.54
     löyty
    -0.53
     anún
    -0.51
     leçons
    -0.51
     liitty
    -0.50
     väli
    -0.50
     énergé
    -0.50
    How
    -0.50
    #
    -0.50
     aikaa
    -0.49
    POSITIVE LOGITS
     you
    1.05
     they
    1.02
     we
    0.91
     those
    0.85
     this
    0.82
    IsMutable
    0.81
     the
    0.79
     your
    0.78
     any
    0.77
    rawDesc
    0.75
    Activations Density 28.333%

    No Known Activations