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. 14316
    Prev
    Next
    INDEX
    Explanations

    terms related to cerebrospinal fluid and its properties or components

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

    fluids or liquids

    np_acts-logits-general · gemini-2.0-flash

    liquid substances and their properties

    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
    SourceChecksum
    -0.84
     Infórmanos
    -0.76
    WaitGroup
    -0.72
     Савезне
    -0.71
    ngdoc
    -0.70
    NameInMap
    -0.68
     تضيفلها
    -0.66
    onViewCreated
    -0.65
    WireFormatLite
    -0.63
    enterOuterAlt
    -0.63
    POSITIVE LOGITS
     poured
    0.54
     mixture
    0.54
     containing
    0.53
     contains
    0.51
     collected
    0.50
     mix
    0.47
     flow
    0.47
     swirl
    0.47
    filled
    0.46
     contain
    0.46
    Activations Density 0.592%

    No Known Activations