© 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. 11178
    Prev
    Next
    INDEX
    Explanations

    elements related to statistical analysis and model evaluation

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

    Tokens after specific keywords like "final", "new", "item"

    np_acts-logits-general · gemini-2.0-flash

    coding and specific terms

    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
     propOrder
    -0.79
     autorytatywna
    -0.68
    BufferException
    -0.63
    ंदीखरीदारी
    -0.62
    ロウィン
    -0.57
    ValueStyle
    -0.57
     Drapeau
    -0.56
    Билгалдахарш
    -0.54
    MLLoader
    -0.54
     الحره
    -0.52
    POSITIVE LOGITS
    TagMode
    0.64
     plutôt
    0.41
     necesite
    0.35
     aseg
    0.34
     naprawdę
    0.34
     limpio
    0.32
     veo
    0.32
     gustó
    0.32
     prévoir
    0.31
    UnusedPrivate
    0.31
    Activations Density 1.035%

    No Known Activations