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. 15778
    Prev
    Next
    INDEX
    Explanations

    phrases emphasizing focus and attention

    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
     Methuen
    -0.79
     Schalke
    -0.71
    Plata
    -0.71
     Pip
    -0.68
     tubing
    -0.67
     Sodom
    -0.65
     Bombay
    -0.64
    Tub
    -0.64
    nb
    -0.64
     employé
    -0.64
    POSITIVE LOGITS
    FOCUS
    1.76
     FOCUS
    1.68
     focus
    1.67
     Focus
    1.63
    Focus
    1.55
     Fokus
    1.48
    focus
    1.47
     focussed
    1.46
     focuses
    1.43
     focussing
    1.43
    Activations Density 0.082%

    No Known Activations