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
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 11
    Prev
    Next
    INDEX
    Explanations

    terms related to competition and ranking

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

    mathematical or scientific terminology related to statistical properties, like "degenerate", "energy states", and quantitative terms.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    concepts related to degeneracy and ties in mathematical or competitive contexts.

    oai_token-act-pair · gpt-4oTriggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     fût
    -0.38
     quartiers
    -0.35
     marrón
    -0.35
    ässä
    -0.33
     bēr
    -0.32
     queſta
    -0.32
    hidup
    -0.32
    vės
    -0.32
     Betreuung
    -0.32
    ARGUMENT
    -0.31
    POSITIVE LOGITS
    ंदीखरीदारी
    0.67
     ब्रेकडाउन
    0.66
    IsMutable
    0.60
    bitField
    0.57
    Iden
    0.46
    tvguidetime
    0.46
    webElementGuid
    0.44
     ویکی‌پدی
    0.43
     ambiguity
    0.43
    ècie
    0.42
    Activations Density 0.239%

    No Known Activations