© 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. 36-GEMMASCOPE-RES-16K
    11. 6138
    Prev
    Next
    INDEX
    Explanations

    phrases related to mathematical calculations, especially those involving determining common factors or performing operations like maximization or minimization

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_36/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.36.hook_resid_post
    Hook Layer
    36
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     böyle
    -0.27
     hayal
    -0.27
     söz
    -0.26
     özellikle
    -0.26
    ispiele
    -0.26
    merkt
    -0.26
     büyük
    -0.25
     kerül
    -0.25
    associazione
    -0.24
     ağır
    -0.24
    POSITIVE LOGITS
     thickness
    1.48
     diameter
    1.38
     length
    1.31
     temperature
    1.30
     width
    1.26
     size
    1.24
     duration
    1.23
     height
    1.23
     density
    1.20
     Thickness
    1.20
    Activations Density 4.633%

    No Known Activations