© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 2977
    Prev
    Next
    INDEX
    Explanations

    terms related to water supply and management

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

    details about water resources, water systems, and water management.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 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
    ModelState
    -0.34
     vayas
    -0.32
     courir
    -0.32
     involu
    -0.32
    Ge
    -0.32
    Loaded
    -0.32
    Py
    -0.31
     George
    -0.31
    zy
    -0.31
     track
    -0.31
    POSITIVE LOGITS
     water
    0.91
     agua
    0.78
     Water
    0.77
     água
    0.76
     aquifers
    0.75
    water
    0.73
     kasarigan
    0.73
     WATER
    0.71
    Water
    0.71
     potable
    0.71
    Activations Density 0.142%

    No Known Activations