© 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. 109414
    Prev
    Next
    INDEX
    Explanations

    sections of text that contain comments, notes, or references

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

    notes and references

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     يتيمه
    -0.86
    Personensuche
    -0.81
    ConstraintMaker
    -0.80
    AddTagHelper
    -0.77
     GenerationType
    -0.74
    انجليز
    -0.72
    Xna
    -0.71
     ModelExpression
    -0.71
    Gön
    -0.70
    OGND
    -0.70
    POSITIVE LOGITS
     Notes
    0.77
    Notes
    0.77
     notes
    0.75
     NOTES
    0.67
     note
    0.67
    notes
    0.65
     Note
    0.64
    Note
    0.64
     comments
    0.63
     Bemerkungen
    0.63
    Activations Density 0.048%

    No Known Activations