© 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-2-2B
    7. Residual Stream - 65k
    9. 20-GEMMASCOPE-RES-65K
    11. 391
    Prev
    Next
    INDEX
    Explanations

    elements related to programming and coding logic

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

    questions related to programming and technical explanations

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

    spanish conjunctions y pero

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     Cæsar
    -1.49
     Monfieur
    -1.48
     Anſ
    -1.45
     Theſe
    -1.39
     avoient
    -1.37
     ―――――
    -1.37
     CHAPITRE
    -1.35
     étoient
    -1.34
     Houſe
    -1.33
     Efq
    -1.33
    POSITIVE LOGITS
     de
    0.78
     en
    0.65
     est
    0.64
     con
    0.60
     es
    0.58
     o
    0.58
     entre
    0.57
     que
    0.57
     ya
    0.57
     la
    0.56
    Activations Density 0.359%

    No Known Activations