© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 3938
    Prev
    Next
    INDEX
    Explanations

    phrases about ways to do something

    oai_token-act-pair · gemini-2.0-flash

    phrases beginning with 'how'

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

    how

    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
     way
    -0.57
    Hochspringen
    -0.54
    Way
    -0.49
    还有一个
    -0.48
     façon
    -0.48
    STP
    -0.48
    way
    -0.48
    Története
    -0.48
    how
    -0.47
    bital
    -0.46
    POSITIVE LOGITS
    LookAnd
    0.83
    UpInside
    0.77
     to
    0.72
    httphttps
    0.71
    +#+#
    0.66
     contextLoads
    0.66
     getItemId
    0.66
    enterOuterAlt
    0.64
     you
    0.63
    RegistryLite
    0.63
    Activations Density 2.584%

    No Known Activations