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. 1410
    Prev
    Next
    INDEX
    Explanations

    terms related to "presence" or "presidential" contexts

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

    Tokens starting with "Pres"

    np_acts-logits-general · gemini-2.0-flash

    words starting with Pres

    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
    gdx
    -0.69
    ỡng
    -0.67
    InputBorder
    -0.66
    坞
    -0.65
     Wikispecies
    -0.64
    ItemBackground
    -0.63
    WithIOException
    -0.62
    ToAction
    -0.62
    ksikon
    -0.61
    etcode
    -0.59
    POSITIVE LOGITS
    Pre
    0.96
     Pre
    0.87
    PRE
    0.83
     PRE
    0.83
    pre
    0.80
     PRES
    0.76
     pre
    0.76
    PRES
    0.74
     Prest
    0.72
     prec
    0.71
    Activations Density 0.356%

    No Known Activations