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. 14644
    Prev
    Next
    INDEX
    Explanations

    specific references and citations within written works

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

    Punctuation and articles

    np_acts-logits-general · gemini-2.0-flash

    titles and headings

    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
     AssemblyCulture
    -1.03
     typelib
    -0.96
     Wikimedijinoj
    -0.87
    ंदीखरीदारी
    -0.86
    WebControls
    -0.86
     ſever
    -0.85
     purpoſe
    -0.85
     photolibrary
    -0.85
     ſche
    -0.82
     hentai
    -0.82
    POSITIVE LOGITS
     The
    1.02
     the
    0.89
     A
    0.83
     Where
    0.79
     a
    0.77
     What
    0.72
     From
    0.70
    The
    0.70
     Who
    0.68
     
    0.68
    Activations Density 0.160%

    No Known Activations