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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 45826
    Prev
    Next
    INDEX
    Explanations

    instances of the word "see" and related references, indicating a focus on observation or awareness

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

    see also

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    NamedQueries
    -0.45
     oog
    -0.35
     tribu
    -0.34
    stridium
    -0.34
    탤
    -0.33
    copyTo
    -0.33
    脚注の使い方
    -0.33
     explored
    -0.33
    gie
    -0.33
    bergement
    -0.33
    POSITIVE LOGITS
     też
    0.80
     also
    0.77
     alſo
    0.64
     también
    0.64
     również
    0.63
     também
    0.62
     επίσης
    0.62
     także
    0.61
     myös
    0.59
     ainfi
    0.59
    Activations Density 0.107%

    No Known Activations