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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 1399
    Prev
    Next
    INDEX
    Explanations

    references to spa and holiday experiences tailored to families and couples

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

    Following "Mr" or "Resorts"

    np_acts-logits-general · gemini-2.0-flash

    resorts and Mr

    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
    BeginInit
    -0.56
    fiddle
    -0.55
     ABN
    -0.54
    Naissance
    -0.53
    RTLR
    -0.52
     jScrollPane
    -0.51
    ora
    -0.50
    DebuggerNonUser
    -0.50
     simples
    -0.49
    mander
    -0.49
    POSITIVE LOGITS
     resort
    0.81
     resorts
    0.72
     Resort
    0.70
    rrggbb
    0.67
    Resort
    0.65
     reps
    0.64
    resort
    0.61
     rep
    0.61
     villaggio
    0.60
     pools
    0.59
    Activations Density 0.044%

    No Known Activations