© 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. 4043
    Prev
    Next
    INDEX
    Explanations

    words near discourse markers like quotes and conversation

    oai_token-act-pair · gemini-2.0-flash

    phrases or quotes

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

    a, the

    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
     frattempo
    -0.40
     kem
    -0.37
     braccia
    -0.36
    uri
    -0.36
    ull
    -0.36
    <
    -0.35
    hoeddwyd
    -0.35
    ries
    -0.34
     grå
    -0.34
     registros
    -0.34
    POSITIVE LOGITS
     phrase
    1.20
     sentence
    1.13
     statement
    1.13
     tagline
    1.10
     quote
    1.08
     slogan
    1.04
     quotation
    0.97
     motto
    0.94
     frase
    0.93
     utterance
    0.93
    Activations Density 6.329%

    No Known Activations