© 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. 24-GEMMASCOPE-TRANSCODER-16K
    7. 15129
    Prev
    Next
    INDEX
    Explanations

    that is hard to tell; it seems to activate on many common words such as "it", "be", "that", "is", and some seemingly random content words.

    oai_token-act-pair · gemini-2.0-flash

    it

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.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
     ModelExpression
    -0.77
     Roskov
    -0.77
     '\\;'
    -0.74
    StructEnd
    -0.70
    RegressionTest
    -0.70
     препратки
    -0.69
    Aholisi
    -0.60
     Drapeau
    -0.60
    Portail
    -0.59
     الحره
    -0.59
    POSITIVE LOGITS
    '
    0.78
    ’
    0.69
    ty
    0.54
    tle
    0.48
     all
    0.45
    ISupport
    0.44
    cy
    0.43
     Большая
    0.43
     indicates
    0.43
    lis
    0.41
    Activations Density 10.744%

    No Known Activations