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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 5679
    Prev
    Next
    INDEX
    Explanations

    words, phrases, and patterns often found in scientific and medical texts, including references, deficiencies, processes, and the interaction of elements.

    oai_token-act-pair · gemini-2.0-flash

    words related to secretion or production

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

    Process of production

    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
     récomp
    -0.51
     findViewById
    -0.51
     tracing
    -0.49
     rápido
    -0.48
     aperto
    -0.48
     Tracer
    -0.47
    findViewById
    -0.46
     Púb
    -0.46
     Tracing
    -0.46
     chiuso
    -0.46
    POSITIVE LOGITS
     secrete
    0.86
     produce
    0.69
     releases
    0.68
    mybatisplus
    0.68
    tagHelperRunner
    0.67
     تضيفلها
    0.66
    ArrowToggle
    0.66
     release
    0.65
     חיצוניים
    0.65
    EDEFAULT
    0.65
    Activations Density 3.195%

    No Known Activations