© 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-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 40180
    Prev
    Next
    INDEX
    Explanations

    legal terminology and constructs related to criminal trials and offenses

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

    Version numbers and code identifiers

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

    The neuron detects structural or symbolic tokens—numbers, single-letter/math symbols, brackets and formatting/LaTeX tokens, and other document-structure markers.

    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
     Goods
    -0.43
    thums
    -0.41
    cho
    -0.40
     goods
    -0.40
    mtrl
    -0.38
     gela
    -0.38
    bosky
    -0.38
    Goods
    -0.38
    ене
    -0.37
     cho
    -0.36
    POSITIVE LOGITS
     AssemblyVersion
    0.56
    請繼續往下閱讀
    0.54
    Tembelea
    0.51
    tvguidetime
    0.50
     HasFactory
    0.47
    posedge
    0.47
     complètes
    0.46
    OGND
    0.46
     nadzieję
    0.45
     Administrativna
    0.44
    Activations Density 0.438%

    No Known Activations