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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 3-GEMMASCOPE-RES-16K
    6. 6190
    Prev
    Next
    INDEX
    Explanations

    technical terminology and references related to scientific studies and concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     sociali
    -0.59
     Tenggara
    -0.58
    lafen
    -0.57
     gouvernements
    -0.56
    Tē
    -0.56
    وفاته
    -0.55
    OrBuilder
    -0.55
     OnInit
    -0.54
     Campionato
    -0.53
     androidx
    -0.53
    POSITIVE LOGITS
     ModelExpression
    0.72
    Aktualisiert
    0.64
    TintMode
    0.58
     Normdatei
    0.58
    makeText
    0.55
     defaultstate
    0.54
    protoc
    0.54
    WebKit
    0.53
    bitField
    0.52
    AutoField
    0.50
    Activations Density 1.324%

    No Known Activations