© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 3397
    Prev
    Next
    INDEX
    Explanations

    quantitative phrases or estimates related to amounts and timelines

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

    numerical phrases involving ranges or approximations (like "five to six", "if not more", "hundreds of thousands").

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

    expressions that convey approximate quantities or ranges, using hedged estimates and alternative bounds.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    phrases expressing uncertainty or approximation, particularly with numerical ranges or estimates (like "if not", "or more", "or less", "maybe a little before or after").

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    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
    writerow
    -0.52
    umą
    -0.42
    ConverterFactory
    -0.42
     denn
    -0.41
    ArgsConstructor
    -0.41
     presumed
    -0.40
     forn
    -0.40
     Obrador
    -0.40
     interface
    -0.39
    manı
    -0.38
    POSITIVE LOGITS
    GEBURTSDATUM
    0.71
     ویکی‌پدی
    0.65
     transfieras
    0.56
     betweenstory
    0.49
    Datuak
    0.48
    EndContext
    0.47
     kasarigan
    0.47
    RegressionTest
    0.46
    complexContent
    0.45
     ivelany
    0.45
    Activations Density 0.350%

    No Known Activations