© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 4481
    Prev
    Next
    INDEX
    Explanations

    sentences talking about feelings

    oai_token-act-pair · gemini-2.0-flash

    the word 'something'

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

    something

    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
    yntaxException
    -0.73
    ftagPool
    -0.57
    EndContext
    -0.52
     IConfiguration
    -0.49
    lapsingToolbar
    -0.49
     ")[
    -0.47
    łaś
    -0.46
    ='\
    -0.45
    reportWebVitals
    -0.44
    lytes
    -0.44
    POSITIVE LOGITS
     something
    3.25
    something
    2.84
    Something
    2.83
     Something
    2.75
     SOMETHING
    2.19
     qualcosa
    2.14
     somethin
    2.06
     algo
    1.85
     quelque
    1.74
     sesuatu
    1.74
    Activations Density 2.791%

    No Known Activations