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. 18-GEMMASCOPE-RES-16K
    6. 14328
    Prev
    Next
    INDEX
    Explanations

    information related to wildfires and their containment efforts

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

    Firefighting, containment, and structures related to fire

    np_acts-logits-general · gemini-2.0-flash

    fire containment and extinguishing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    XmlAccessType
    -0.74
     الحره
    -0.69
    DoubleQuotes
    -0.63
    AutoField
    -0.63
    WebMethod
    -0.58
    MigrationBuilder
    -0.58
     CURIAM
    -0.57
    MenuGroup
    -0.56
    GetKeyDown
    -0.55
    jahtera
    -0.54
    POSITIVE LOGITS
     fire
    1.24
     firefighting
    1.19
     extinguishing
    1.07
     fires
    1.05
     Fire
    1.04
     firefighters
    1.02
    Fire
    0.98
     extingu
    0.97
     firefighter
    0.96
     extinguish
    0.95
    Activations Density 0.079%

    No Known Activations