© 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-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 44751
    Prev
    Next
    INDEX
    Explanations

    connections between various elements in a comparative or evaluative context

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

    different post-processing

    np_acts-logits-general · gemini-2.5-flash-lite

    the neuron responds to salient topical content words (important technical or academic terms) in the document.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ///</
    -0.41
    providedIn
    -0.38
    ⓧ
    -0.36
    issauga
    -0.34
     autorytatywna
    -0.33
     capot
    -0.33
     ujednoznacz
    -0.33
    SharedCtor
    -0.32
     joão
    -0.32
    āla
    -0.32
    POSITIVE LOGITS
     useAppContext
    0.50
     queſta
    0.50
    数料
    0.50
    Kjelder
    0.49
     ProtoMessage
    0.48
     mxArray
    0.46
     للمعارف
    0.46
    anzi
    0.46
     Administrativna
    0.45
    frameCount
    0.44
    Activations Density 0.329%

    No Known Activations