© 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. 13446
    Prev
    Next
    INDEX
    Explanations

    a grab bag of specific patterns in code, French words, and personal ads

    oai_token-act-pair · gemini-2.0-flash

    words related to specificity or plurality

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

    Code or foreign languages

    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
    writeFieldEnd
    -0.81
    EndContext
    -0.77
    oredCriteria
    -0.76
    IUrlHelper
    -0.69
    StructEnd
    -0.68
    SharedDtor
    -0.67
     ciasc
    -0.67
    SequentialGroup
    -0.67
    DeleteBehavior
    -0.66
    DebuggerNonUser
    -0.66
    POSITIVE LOGITS
     those
    0.94
    Those
    0.81
    those
    0.80
     Those
    0.79
     mga
    0.74
    那些
    0.74
     ones
    0.73
     những
    0.67
     THOSE
    0.60
    ские
    0.60
    Activations Density 0.826%

    No Known Activations