© 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. 11549
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to studies about weight loss or medical interventions and choices

    oai_token-act-pair · gemini-2.0-flash

    terms associated with weight loss

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

    Losing

    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
     actionMode
    -0.60
    üstet
    -0.46
    thalten
    -0.45
    jooq
    -0.44
     armées
    -0.44
    hamshire
    -0.43
     armée
    -0.43
     springfox
    -0.42
     cinéma
    -0.42
    dje
    -0.42
    POSITIVE LOGITS
     loss
    2.19
     lost
    2.16
     lose
    2.14
     losing
    2.13
     loses
    1.99
    loss
    1.91
     Losing
    1.91
    Losing
    1.89
    Loss
    1.88
     Loss
    1.86
    Activations Density 0.965%

    No Known Activations