© 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. 15653
    Prev
    Next
    INDEX
    Explanations

    verbs, especially "be" and "-ing" verbs, as well as related words like "can" and "should"

    oai_token-act-pair · gemini-2.0-flash

    letter prefixes

    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
     jScrollPane
    -0.66
     يتيمه
    -0.65
    expandindo
    -0.60
     calendriers
    -0.55
     quadr
    -0.54
    Gweler
    -0.52
     roux
    -0.52
     resourceCulture
    -0.52
    Cyfeiriadau
    -0.52
    lossians
    -0.52
    POSITIVE LOGITS
     de
    0.86
     ex
    0.83
     A
    0.75
     ab
    0.72
    a
    0.70
     а
    0.67
     à
    0.61
    A
    0.60
     á
    0.58
    aA
    0.58
    Activations Density 7.072%

    No Known Activations