© 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. 7397
    Prev
    Next
    INDEX
    Explanations

    sentences that mention briefings and secret information

    oai_token-act-pair · gemini-2.0-flash

    words related to secrecy or classification

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

    Classified information

    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
     profesión
    -0.54
    ulturen
    -0.53
     laborales
    -0.51
     judiciales
    -0.50
    存于互联网档案馆
    -0.50
    esgue
    -0.49
     campagna
    -0.49
    InstanceState
    -0.48
     vectoriales
    -0.48
    mybatisplus
    -0.48
    POSITIVE LOGITS
     classified
    1.02
     top
    1.00
    classified
    0.98
     secret
    0.96
    top
    0.93
    secret
    0.87
     Top
    0.84
    TOP
    0.84
    Top
    0.82
     TOP
    0.81
    Activations Density 1.979%

    No Known Activations