© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 10227
    Prev
    Next
    INDEX
    Explanations

    references to statistical relationships and effects in data analysis

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

    mathematical or scientific statements that emphasize comparative or contrasting relationships, numerical values, directions, and polarity (such as positive/negative, higher/lower, increase/decrease, directionality, or contrast between groups or conditions).

    oai_token-act-pair · gpt-4.1-2025-04-14Triggered by @ywhp2c8tjm

    words indicating positive or negative value or direction, often in the context of scientific or mathematical data.

    oai_token-act-pair · gemini-2.0-flashTriggered by @ywhp2c8tjm

    Positive or negative changes, often expressed numerically

    np_acts-logits-general · gemini-2.0-flash

    up or down

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    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
    transQ
    -0.63
     referrerpolicy
    -0.59
    writeFieldEnd
    -0.58
     OFDb
    -0.57
     فريبيس
    -0.57
    featureID
    -0.54
     Meksiku
    -0.54
    󠁴
    -0.53
     franchi
    -0.53
    CppMethod
    -0.53
    POSITIVE LOGITS
     increase
    0.41
    posedge
    0.39
     افزایش
    0.36
     increases
    0.35
     increasing
    0.35
     dispuesto
    0.34
     entgeg
    0.34
    increase
    0.33
     upward
    0.32
     INCREASE
    0.32
    Activations Density 0.954%

    No Known Activations