© 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. 7-GEMMASCOPE-TRANSCODER-16K
    7. 10045
    Prev
    Next
    INDEX
    Explanations

    instances of accusations of crimes, particularly accusations of rape and harming the innocent

    oai_token-act-pair · gemini-2.0-flash

    accusations and guilt

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_7/width_16k/average_l0_70
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.7.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
    utnik
    -0.61
    الحياه
    -0.57
     дописавши
    -0.53
    Šaltiniai
    -0.51
    èdia
    -0.50
     handleMessage
    -0.50
    setCancelable
    -0.47
    aarrggbb
    -0.46
    SignUp
    -0.46
     eventdata
    -0.46
    POSITIVE LOGITS
    ſelf
    0.71
     Majefty
    0.70
     raiſ
    0.68
     againſt
    0.68
     itſelf
    0.66
     ſtre
    0.64
     accusations
    0.62
     Theſe
    0.60
     allegations
    0.59
     leſs
    0.58
    Activations Density 1.233%

    No Known Activations