© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 38-GEMMASCOPE-RES-16K
    6. 13621
    Prev
    Next
    INDEX
    Explanations

    sequences of player names and their performance metrics in sports

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Harwood
    -0.37
     Architektur
    -0.34
    OrBuilder
    -0.34
     onCancelled
    -0.34
     orgánico
    -0.34
    łoży
    -0.33
     arquitetura
    -0.33
     arkite
    -0.32
     arquitectura
    -0.32
     poivre
    -0.32
    POSITIVE LOGITS
     receivers
    1.06
     receiver
    1.01
     receptions
    0.94
     reception
    0.91
     Receiver
    0.86
     RECE
    0.85
     routes
    0.84
     Reception
    0.83
     receptor
    0.83
     recep
    0.81
    Activations Density 0.195%

    No Known Activations