© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 14635
    Prev
    Next
    INDEX
    Explanations

    cryptographic terms, memory addresses, and related constants

    oai_token-act-pair · gemini-2.0-flash

    software and networking contexts

    np_max-act-logits · gemini-2.0-flash

    numbers

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109

    digits

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    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
     Six
    -0.57
    2
    -0.55
    dropIfExists
    -0.55
    1
    -0.54
     Nine
    -0.53
    Six
    -0.53
     Five
    -0.52
    -0.52
     six
    -0.52
    4
    -0.52
    POSITIVE LOGITS
     ویکی‌آمباردا
    0.86
    ]")]
    0.79
    aarrggbb
    0.78
    ագրություններ
    0.76
    extAlignment
    0.74
    UnsafeEnabled
    0.68
    SEGUIR
    0.67
     pageable
    0.66
    فایل‌لار
    0.65
     فريبيس
    0.65
    Activations Density 1.647%

    No Known Activations