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. 20-GEMMASCOPE-RES-16K
    6. 15530
    Prev
    Next
    INDEX
    Explanations

    elements related to configuration settings and routing in a programming context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Code related to registering/registration

    np_acts-logits-general · gemini-2.0-flash

    register

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    badlogic
    -0.52
    yntaxException
    -0.52
     hObject
    -0.50
     arXiv
    -0.45
    optarg
    -0.45
    variates
    -0.45
    ClientSize
    -0.44
     ControllerBase
    -0.42
     énergétique
    -0.42
    ruptedException
    -0.42
    POSITIVE LOGITS
     register
    0.96
    register
    0.95
     registering
    0.92
     Register
    0.91
    Register
    0.84
     registration
    0.84
     registered
    0.82
     注册
    0.80
     REGISTER
    0.77
     registrations
    0.76
    Activations Density 0.403%

    No Known Activations