© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 15488
    Prev
    Next
    INDEX
    Explanations

    words related to causation and conditional relationships, often with a negative or problematic connotation.

    oai_token-act-pair · gemini-2.0-flash

    positive adjectives

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

    positive adjectives

    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
    Negative Logits
    nefs
    -0.77
    neſs
    -0.71
     pleaſure
    -0.70
     photolibrary
    -0.70
    出版年
    -0.68
     chofe
    -0.67
    argout
    -0.67
     ſche
    -0.67
     tramontina
    -0.65
    InjectAttribute
    -0.65
    POSITIVE LOGITS
     great
    0.90
     nice
    0.87
     interesting
    0.85
     good
    0.84
     wonderful
    0.82
     fantastic
    0.81
     excellent
    0.79
     lovely
    0.78
     fabulous
    0.77
     terrific
    0.75
    Activations Density 1.960%

    No Known Activations