© 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-2-2B
    4. Residual Stream - 16k
    5. 16-GEMMASCOPE-RES-16K
    6. 8743
    Prev
    Next
    INDEX
    Explanations

    instances of the word "gap" and its various contexts or implications

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

    "gap" and its context

    np_acts-logits-general · gemini-2.0-flash

    gap between

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     cheminée
    -0.65
     otomatig
    -0.61
     nucléaire
    -0.59
    maketitle
    -0.59
    anyahu
    -0.59
     Wyman
    -0.58
     فريبيس
    -0.58
     CURIAM
    -0.57
     Hogarth
    -0.56
     henvisninger
    -0.55
    POSITIVE LOGITS
     gap
    1.28
     between
    1.21
     Gap
    1.16
     gaps
    1.10
    Gap
    1.10
    gap
    1.07
     Gaps
    1.03
    between
    1.02
     BETWEEN
    0.94
     tussen
    0.92
    Activations Density 0.458%

    No Known Activations