Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 11008
    Prev
    Next
    INDEX
    Explanations

    symbols and numerical representations, particularly those related to dates and quantities

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

    "to" followed by a verb/adjective

    np_acts-logits-general · gemini-2.0-flash

    time ranges and durations

    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
     disambiguazione
    -0.77
    +#+#
    -0.70
    homonymie
    -0.68
    WriteTagHelper
    -0.65
     дописавши
    -0.58
    مراجع
    -0.58
    HostException
    -0.58
     Мексичка
    -0.58
     Hunan
    -0.57
    XtraReports
    -0.57
    POSITIVE LOGITS
     EClass
    0.47
    OMI
    0.46
     half
    0.45
     siglos
    0.42
     Merritt
    0.42
    塵
    0.40
    sites
    0.39
     Telerik
    0.39
    half
    0.39
    学期
    0.39
    Activations Density 0.348%

    No Known Activations