© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 1-GEMMASCOPE-RES-16K
    6. 438
    Prev
    Next
    INDEX
    Explanations

    the repeated use of the term "par," which may signify parameters or levels in a statistical or evaluative context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    :✨
    -0.43
    StructEnd
    -0.38
    WireFormatLite
    -0.38
    Autoritní
    -0.36
     œil
    -0.35
    WebElementEntity
    -0.35
     Keuangan
    -0.33
    thinkable
    -0.33
    __*/
    -0.32
     Савезне
    -0.32
    POSITIVE LOGITS
     PAR
    2.91
     Par
    2.84
    Par
    2.81
    par
    2.78
    PAR
    2.73
     par
    2.70
     Pars
    2.06
    Pars
    2.05
     Parrish
    1.99
     پار
    1.97
    Activations Density 0.221%

    No Known Activations