© 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. 4606
    Prev
    Next
    INDEX
    Explanations

    negative sentiment, as shown by words like "not", "never", and negative contractions.

    oai_token-act-pair · gemini-2.0-flash

    the word 'not'

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

    suggests negations

    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
    ніципа
    -0.82
    SourceChecksum
    -0.79
     يتيمه
    -0.78
    IonicModule
    -0.73
    WebVitals
    -0.73
    Personensuche
    -0.71
    adaptiveStyles
    -0.70
     Вікі
    -0.69
    CloseOperation
    -0.69
     betweenstory
    -0.68
    POSITIVE LOGITS
    inaire
    0.46
    Климат
    0.43
     onView
    0.41
     glos
    0.40
     syllabus
    0.40
     compromise
    0.40
     kolej
    0.39
    ソ
    0.39
     predicted
    0.39
     nhiêu
    0.38
    Activations Density 0.834%

    No Known Activations