© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 35-GEMMASCOPE-RES-16K
    11. 5560
    Prev
    Next
    INDEX
    Explanations

    references to physical interactions or activities that are considered inappropriate or unacceptable

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.35.hook_resid_post
    Hook Layer
    35
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ური
    -0.68
    pyplot
    -0.56
    újo
    -0.54
     informée
    -0.54
    ArrowToggle
    -0.54
     nahilalakip
    -0.54
    комму
    -0.52
     consig
    -0.50
    haël
    -0.50
    Identyfik
    -0.48
    POSITIVE LOGITS
     nemico
    0.46
    BagConstraints
    0.45
     gravedad
    0.43
    SBATCH
    0.42
     hospod
    0.39
     repuesto
    0.38
     igualdad
    0.38
     utilidad
    0.38
     consommate
    0.38
    Egypte
    0.37
    Activations Density 0.674%

    No Known Activations