© 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. 4735
    Prev
    Next
    INDEX
    Explanations

    the word "incident" and words that can have negative connotations such as "encounter" and "instance"

    oai_token-act-pair · gemini-2.0-flash

    incident

    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
     myſelf
    -2.03
     Efq
    -1.83
     Monfieur
    -1.69
     raiſ
    -1.67
     himſelf
    -1.67
     auffi
    -1.60
     purpoſe
    -1.59
     themſelves
    -1.59
     ſmall
    -1.58
     uſed
    -1.58
    POSITIVE LOGITS
    ↵
    1.04
     
    0.96
      
    0.90
    <eos>
    0.85
     I
    0.83
    .
    0.83
    L
    0.78
    '
    0.75
    er
    0.75
    0
    0.74
    Activations Density 1.430%

    No Known Activations