© 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. 0-GEMMASCOPE-TRANSCODER-16K
    7. 14586
    Prev
    Next
    INDEX
    Explanations

    verbs related to speech and acknowledgement

    oai_token-act-pair · gemini-2.0-flash

    admitted

    np_max-act · gemini-2.0-flash

    admitted

    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_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
    Not in Any Lists

    No Comments

    Negative Logits
    RefreshLayout
    -0.47
     vayan
    -0.45
    ■■
    -0.44
     finnas
    -0.42
     Tengo
    -0.42
    PageRoute
    -0.41
     olisi
    -0.39
    ArgumentParser
    -0.39
     Cruises
    -0.38
    haben
    -0.37
    POSITIVE LOGITS
    <bos>
    0.91
     Мексичка
    0.83
    EDEFAULT
    0.82
    httphttps
    0.79
    awtextra
    0.79
    istoitu
    0.76
    enterOuterAlt
    0.74
    UnusedPrivate
    0.72
    twimg
    0.72
     <=",
    0.72
    Activations Density 2.147%

    No Known Activations