Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 21788
    Prev
    Next
    INDEX
    Explanations

    surface-related scientific terms in the context of material properties and structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     paste
    -0.24
    glGen
    -0.22
    <eos>
    -0.22
     Sachsen
    -0.22
     document
    -0.22
    ufficio
    -0.22
     دفتر
    -0.22
    anniversaire
    -0.21
     Vector
    -0.20
     formula
    -0.20
    POSITIVE LOGITS
    出版年
    0.90
    RegressionTest
    0.88
    OGND
    0.81
    GEBURTSDATUM
    0.79
    NameInMap
    0.77
    RectangleBorder
    0.75
     tartalomajánló
    0.75
     виправивши
    0.75
    posedge
    0.74
     nahilalakip
    0.73
    Activations Density 0.081%

    No Known Activations