© 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. 4152
    Prev
    Next
    INDEX
    Explanations

    the word "weapon" or related words, sometimes with a small activation related to a person

    oai_token-act-pair · gemini-2.0-flash

    weapon

    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
     
    -0.69
     da
    -0.61
    Programme
    -0.54
    ↵↵
    -0.54
     man
    -0.54
     es
    -0.53
     d
    -0.49
    Ỉ
    -0.49
     the
    -0.47
     de
    -0.47
    POSITIVE LOGITS
     Monfieur
    1.19
     Efq
    1.15
     myſelf
    1.10
     uſed
    1.04
     ſtate
    1.01
     Majefty
    1.00
     juſ
    0.99
     Anſ
    0.99
     Jefus
    0.98
     purpoſe
    0.98
    Activations Density 3.165%

    No Known Activations