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. 15550
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to blocking nominations or judicial appointments in a political context

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

    appointments or nominations

    np_acts-logits-general · gemini-2.0-flash

    appointment confirmation

    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
    FormTagHelper
    -0.55
     AssemblyCulture
    -0.55
    tagHelperRunner
    -0.50
    !("{}",
    -0.50
    CSR
    -0.49
    luiting
    -0.46
    ='_
    -0.46
    
    -0.45
     beth
    -0.44
     '*',
    -0.43
    POSITIVE LOGITS
     appointment
    1.59
     appoint
    1.55
     appointed
    1.50
     appointments
    1.43
     appointing
    1.43
     APPOINT
    1.36
     Appointments
    1.36
     Appointment
    1.35
     appoints
    1.33
     appointe
    1.32
    Activations Density 0.398%

    No Known Activations