© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 74240
    Prev
    Next
    INDEX
    Explanations

    quantifiable measurements and frequencies associated with events

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

    twice, per, group

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

    The neuron activates at tokens that mark or begin the model's response (i.e., the start of the assistant's turn/answer).

    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
    TagMode
    -0.43
    Citiți
    -0.43
    rrggbb
    -0.36
    XmlEnum
    -0.36
     finally
    -0.35
    SourceChecksum
    -0.34
    XmlAccessorType
    -0.34
     enfin
    -0.34
    Vaata
    -0.33
    ConstraintMaker
    -0.32
    POSITIVE LOGITS
    niſſe
    0.62
     kasarigan
    0.53
    fender
    0.48
     utafitiHapana
    0.48
     فريبيس
    0.47
    __.__
    0.46
    polation
    0.46
     chloroform
    0.46
    Pop
    0.45
     الاطلاع
    0.45
    Activations Density 0.001%

    No Known Activations