© 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. 0-GEMMASCOPE-TRANSCODER-16K
    7. 9693
    Prev
    Next
    INDEX
    Explanations

    the word "toxicity"

    oai_token-act-pair · gemini-2.0-flash

    toxicity

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.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
     Monfieur
    -2.02
     Efq
    -1.91
     itſelf
    -1.86
     myſelf
    -1.83
     Diſ
    -1.80
     Majefty
    -1.79
     Eſ
    -1.78
     purpoſe
    -1.74
     Theſe
    -1.73
     Houſe
    -1.72
    POSITIVE LOGITS
    ↵↵
    1.98
     
    1.09
     I
    1.07
     "
    1.03
     A
    0.99
     (
    0.95
     '
    0.92
    .
    0.92
     -
    0.89
     S
    0.88
    Activations Density 0.185%

    No Known Activations