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. 10-GEMMASCOPE-RES-16K
    6. 2862
    Prev
    Next
    INDEX
    Explanations

    statistical assumptions or hypotheses related to data and its properties

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ex
    -0.47
    wpi
    -0.47
    却
    -0.45
    testng
    -0.44
    لمة
    -0.44
     native
    -0.43
    GEBURTS
    -0.43
    beau
    -0.42
    KeyPair
    -0.42
     nonatomic
    -0.41
    POSITIVE LOGITS
    featureID
    0.69
    ########.
    0.69
     propOrder
    0.63
    sidemargin
    0.62
     */
    
    
    0.58
    assumed
    0.58
     Assumptions
    0.57
    WebElementEntity
    0.57
     assume
    0.57
     Assumption
    0.56
    Activations Density 0.611%

    No Known Activations