© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 4059
    Prev
    Next
    INDEX
    Explanations

    phrases expressing a lack of something or difficulty

    oai_token-act-pair · gemini-2.0-flash

    say "a, much"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
     obstante
    -0.63
     Quite
    -0.61
     quelconque
    -0.61
    Quite
    -0.57
    expandindo
    -0.56
    riwal
    -0.55
    ()])
    -0.54
    ⓘ
    -0.52
     continúas
    -0.52
    脚注の使い方
    -0.51
    POSITIVE LOGITS
     many
    1.50
     muchas
    1.24
    many
    1.23
     much
    1.23
    Many
    1.20
     Many
    1.18
     muchos
    1.13
     molte
    1.09
     muitos
    1.06
     molti
    1.05
    Activations Density 1.740%

    No Known Activations