© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 3016
    Prev
    Next
    INDEX
    Explanations

    references to legal proceedings and well-known people's thoughts or statements

    oai_token-act-pair · gemini-2.0-flash

    observe

    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
    Negative Logits
     says
    -0.79
     Says
    -0.71
    PerformLayout
    -0.70
    Says
    -0.68
    says
    -0.66
     say
    -0.65
     saith
    -0.63
     SAYS
    -0.62
     Say
    -0.59
    say
    -0.57
    POSITIVE LOGITS
     observed
    1.25
     observe
    1.23
     observation
    1.22
     observing
    1.20
     observations
    1.09
    observed
    1.09
     observes
    1.06
     Observing
    1.05
     Observe
    1.05
    observ
    1.04
    Activations Density 3.249%

    No Known Activations