© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 30-GEMMASCOPE-RES-16K
    11. 13487
    Prev
    Next
    INDEX
    Explanations

    various types and categories of "sort" along with their relations in contexts such as governance or classifications related to noble figures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    following "Lord" or "sort"

    np_acts-logits-general · gemini-2.0-flash

    sort and lord

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_30/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.30.hook_resid_post
    Hook Layer
    30
    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
    -0.73
     Beagle
    -0.55
    Mej
    -0.54
    "}")
    -0.52
     stringBuilder
    -0.52
    ურ
    -0.51
    Jac
    -0.51
    Биография
    -0.51
     Dynamite
    -0.51
     Manu
    -0.50
    POSITIVE LOGITS
    ror
    0.40
    ors
    0.38
    oro
    0.38
    ros
    0.37
    rer
    0.36
    als
    0.36
    HasForeignKey
    0.35
    lrrr
    0.35
    land
    0.35
     delito
    0.34
    Activations Density 0.794%

    No Known Activations