© 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. 19-GEMMASCOPE-TRANSCODER-16K
    7. 7249
    Prev
    Next
    INDEX
    Explanations

    people entering or leaving a building

    oai_token-act-pair · gemini-2.0-flash

    say "door"

    np_max-act-logits · gemini-2.0-flash

    the Method used: 1. Reason: repeated "the" tokens

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_19/width_16k/average_l0_12
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.19.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
    ритори
    -0.50
    ThemeOverlay
    -0.47
    tld
    -0.46
     TAMBÉM
    -0.46
     Luar
    -0.46
     hashtag
    -0.45
     TRABAJO
    -0.45
     hashtags
    -0.44
     Placeholder
    -0.44
    ropolitan
    -0.44
    POSITIVE LOGITS
     door
    2.67
     doors
    2.39
    door
    2.22
     Door
    2.14
    Door
    2.13
     DOOR
    2.08
    doors
    1.96
     Doors
    1.94
     puerta
    1.92
     gates
    1.84
    Activations Density 0.621%

    No Known Activations