© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 5191
    Prev
    Next
    INDEX
    Explanations

    mentions of Manchester United and its players along with other soccer/football terms

    oai_token-act-pair · gemini-2.0-flash

    conjunctions and coordinating phrases

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

    sports and events

    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
    OGND
    -0.93
    AnchorStyles
    -0.81
    stdc
    -0.76
    InitVars
    -0.68
    djangoproject
    -0.68
     onOptions
    -0.67
    IsContent
    -0.66
    WebVitals
    -0.65
     CreateTagHelper
    -0.62
    tagext
    -0.61
    POSITIVE LOGITS
     but
    0.58
     which
    0.49
     kvar
    0.46
    inks
    0.46
     recevez
    0.45
    inkt
    0.44
     lakini
    0.42
    ink
    0.42
     encontraban
    0.41
    Thrown
    0.41
    Activations Density 2.169%

    No Known Activations