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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 8849
    Prev
    Next
    INDEX
    Explanations

    a series of seeming unrelated terms including acronyms (WMS, WFS, WCS, DDT, ECOA, SIG), parts of names (Gian), and common words (the, fair, had, is).

    oai_token-act-pair · gemini-2.0-flash

    programming-related terms or identifiers

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

    code-like text

    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
     CreateTagHelper
    -0.58
    Portály
    -0.54
    IVEREF
    -0.53
    Spoljašnje
    -0.50
    FTFY
    -0.49
    oq
    -0.49
    })));
    -0.48
     attacker
    -0.47
    Karriere
    -0.47
    singing
    -0.45
    POSITIVE LOGITS
     propOrder
    0.78
    )(((
    0.65
    testcase
    0.64
    Geplaatst
    0.64
    JAKARTA
    0.63
    onViewCreated
    0.63
     invokingState
    0.61
    jandro
    0.60
     preocupes
    0.60
     незавершена
    0.59
    Activations Density 3.631%

    No Known Activations