© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 3624
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "the" and associated negations within a context

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

    sole existence

    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
    fiques
    -0.38
    つつ
    -0.36
    getInputStream
    -0.35
    たくさん
    -0.35
    abilité
    -0.35
    לה
    -0.35
    pierw
    -0.34
     a
    -0.34
    chod
    -0.34
     Part
    -0.34
    POSITIVE LOGITS
     един
    1.60
     únicos
    1.60
     único
    1.58
    唯一
    1.56
     únicas
    1.52
     única
    1.51
    唯一的
    1.36
     only
    1.35
     jedin
    1.35
     eneste
    1.34
    Activations Density 0.204%

    No Known Activations