© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 0-GEMMASCOPE-TRANSCODER-16K
    7. 14017
    Prev
    Next
    INDEX
    Explanations

    instances of the word "shit" and words that relate to quantity

    oai_token-act-pair · gemini-2.0-flash

    shit

    np_max-act · gemini-2.0-flash

    shit

    np_max-act-logits · gpt-5-mini
    Triggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     contente
    -0.61
     softens
    -0.60
     tonel
    -0.58
     genomen
    -0.58
    djangoproject
    -0.57
     gebracht
    -0.57
    ↵↵
    -0.57
     equator
    -0.56
     Unidos
    -0.55
    pazität
    -0.55
    POSITIVE LOGITS
     ―――――
    1.13
     Chwiliwch
    0.95
    ValueStyle
    0.86
     ſind
    0.86
    NameInMap
    0.85
     disambiguazione
    0.85
    KURZBESCHREIBUNG
    0.85
    فایل‌لار
    0.83
     Reſ
    0.83
     ――――――――
    0.81
    Activations Density 3.697%

    No Known Activations