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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 14293
    Prev
    Next
    INDEX
    Explanations

    phrases that emphasize exceeding basic expectations or functions

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

    after specific nouns ("Elections," "Spectrum," etc.)

    np_acts-logits-general · gemini-2.0-flash

    beyond the traditional

    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
    ArgsConstructor
    -0.49
    ActionCreators
    -0.49
    CreateModel
    -0.47
    tubers
    -0.44
    AddField
    -0.44
     HomeComponent
    -0.43
    InstrumentedTest
    -0.43
    legi
    -0.42
    nete
    -0.41
     partout
    -0.41
    POSITIVE LOGITS
    principalColumn
    0.44
     intptr
    0.39
    ofluor
    0.37
     CanadaChoose
    0.36
    Portály
    0.36
     FetchType
    0.35
     iconLine
    0.35
     CommonModule
    0.34
    tagHelperRunner
    0.34
     चीज़ों
    0.34
    Activations Density 0.264%

    No Known Activations