© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 7228
    Prev
    Next
    INDEX
    Explanations

    words and phrases used in political or legal discourse, possibly indicating arguments, accusations, or denunciations.

    oai_token-act-pair · gemini-2.0-flash

    terms related to accusations or charges

    np_token-act-pair-logits · gpt-4o-mini

    charges/attack

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.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
     jsPsych
    -0.51
     conscious
    -0.45
    δω
    -0.43
     biblia
    -0.40
    JTextField
    -0.39
     racional
    -0.39
    جاه
    -0.39
    amethasone
    -0.38
    WithIOException
    -0.38
    ൊ
    -0.38
    POSITIVE LOGITS
     charges
    1.25
     charge
    1.24
     charged
    1.13
    charge
    1.12
    charges
    1.10
     attack
    1.06
     charging
    1.06
     Charges
    1.05
    Charge
    1.03
     Charged
    1.03
    Activations Density 1.222%

    No Known Activations