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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 15124
    Prev
    Next
    INDEX
    Explanations

    technical terminology related to app development and software engineering

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

    that describe something

    np_acts-logits-general · gemini-2.0-flash

    hints, motives, businesses, dog

    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
    MigrationBuilder
    -0.57
     kasarigan
    -0.57
    qtype
    -0.56
    󠁴
    -0.54
    tagext
    -0.54
     ***!
    -0.52
    ẖ
    -0.52
    testnet
    -0.50
    лтамалар
    -0.49
     AssemblyCulture
    -0.48
    POSITIVE LOGITS
     étoient
    0.46
     eenige
    0.41
    InstrumentedTest
    0.40
     regeringen
    0.38
    zamos
    0.37
     normaux
    0.36
     giù
    0.35
     burbujas
    0.35
     zijne
    0.34
     occhiali
    0.34
    Activations Density 0.411%

    No Known Activations