© 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. 6917
    Prev
    Next
    INDEX
    Explanations

    political text, including names of political parties, politicians, and related vocabulary

    oai_token-act-pair · gemini-2.0-flash

    the

    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
    спа
    -0.42
    
    -0.42
    näm
    -0.42
    allowNull
    -0.42
     espin
    -0.41
    jot
    -0.40
    yson
    -0.40
    CodeAttribute
    -0.40
     legyen
    -0.40
     pre
    -0.39
    POSITIVE LOGITS
     سكانية
    0.83
     pecuniary
    0.77
     Moslem
    0.75
     negroes
    0.72
     Negroes
    0.71
     Whigs
    0.71
     Opium
    0.70
     препратки
    0.69
     BoxFit
    0.69
     Hän
    0.68
    Activations Density 3.080%

    No Known Activations