© 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. 322
    Prev
    Next
    INDEX
    Explanations

    words related to describing or identifying things, particularly in scientific or analytical contexts. The neuron activates strongly for words like "identified", "observed", "described", "prepared", "introduced", etc. These terms are commonly used when reporting findings, methods, or results in research or technical writing.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @myinzi123

    terms related to scientific measurements and observations

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

    Tokens preceding nouns or results

    np_acts-logits-general · gemini-2.0-flash

    describing research findings

    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
     auroit
    -0.70
     ainfi
    -0.69
     pouvoit
    -0.65
     feroit
    -0.63
    了嗎
    -0.63
    majánló
    -0.62
     houſe
    -0.60
     ſon
    -0.60
     itſelf
    -0.60
     zoude
    -0.59
    POSITIVE LOGITS
     ongoing
    0.54
     Ongoing
    0.47
    된
    0.47
    Displayed
    0.45
    Ongoing
    0.44
     pesky
    0.44
     bulunan
    0.43
    ongoing
    0.42
     internal
    0.41
    Remaining
    0.41
    Activations Density 0.499%

    No Known Activations