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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 9013
    Prev
    Next
    INDEX
    Explanations

    references to military installations and environmental concerns

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

    Locations: bases, sites, or campuses

    np_acts-logits-general · gemini-2.0-flash

    specific locations or sites

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    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
    TextBoxColumn
    -0.40
    XmlAccessType
    -0.38
    "{\
    -0.38
     Guide
    -0.38
    Guide
    -0.37
    spiegel
    -0.36
    itrine
    -0.36
    anny
    -0.35
    BufferedReader
    -0.35
     pun
    -0.35
    POSITIVE LOGITS
    TypedDataSet
    0.70
     campus
    0.60
     campuses
    0.59
     complexo
    0.58
     complex
    0.57
     complejo
    0.57
    éroport
    0.57
     Gelände
    0.57
     grounds
    0.56
    gameserver
    0.56
    Activations Density 0.218%

    No Known Activations