© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 15320
    Prev
    Next
    INDEX
    Explanations

    seems to activate on narrative or descriptive text of a scientific nature, possibly in the area of medicine or geography

    oai_token-act-pair · gemini-2.0-flash

    technical documentation excerpts

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
     consists
    -1.27
     consisted
    -1.24
     consist
    -1.21
     consisting
    -1.05
     consiste
    -0.91
     består
    -0.81
     besteht
    -0.71
     состоит
    -0.69
     Consist
    -0.68
     consistente
    -0.65
    POSITIVE LOGITS
    ">—
    0.76
    ThroughAttribute
    0.61
    InitVars
    0.58
    Abitanti
    0.57
    awtextra
    0.54
    richTextPanel
    0.50
    FieldNumber
    0.49
    figer
    0.49
    GroupLayout
    0.49
     Италијани
    0.49
    Activations Density 1.543%

    No Known Activations