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. 14-GEMMASCOPE-RES-16K
    6. 172
    Prev
    Next
    INDEX
    Explanations

    the verb "stick" and its variations, indicating a preference for staying with or adhering to something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.14.hook_resid_post
    Hook Layer
    14
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ValueStyle
    -0.95
     שוליים
    -0.88
    UTERS
    -0.79
     Olms
    -0.76
    τουργ
    -0.75
    imageio
    -0.74
    endphp
    -0.72
    MessageOf
    -0.71
    virons
    -0.71
    InSection
    -0.70
    POSITIVE LOGITS
     stick
    1.67
     STICK
    1.52
     Stick
    1.51
     Sticks
    1.46
     sticks
    1.46
    stick
    1.39
    STICK
    1.39
    Stick
    1.39
     sticking
    1.31
     stuck
    1.30
    Activations Density 0.088%

    No Known Activations