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. 21-GEMMASCOPE-RES-16K
    6. 15056
    Prev
    Next
    INDEX
    Explanations

    instances of the word "stick" and its variations, indicating discussions about sticking with something or remaining attached to an idea or concept

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Italijanski
    -0.82
     Meksiku
    -0.73
    ArrowToggle
    -0.66
    ----</
    -0.61
     muualla
    -0.60
     שוליים
    -0.59
     חיצוניים
    -0.59
     BoxDecoration
    -0.59
    IUrlHelper
    -0.58
    ChildScrollView
    -0.58
    POSITIVE LOGITS
     stick
    1.15
     Stick
    1.12
    Stick
    1.06
    stick
    1.05
     STICK
    1.01
     sticking
    1.00
     sticks
    0.99
     Sticks
    0.92
    STICK
    0.91
    sticks
    0.83
    Activations Density 0.047%

    No Known Activations