© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 6600
    Prev
    Next
    INDEX
    Explanations

    terms related to specific purposes and roles in various contexts

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

    complaints or problems

    np_acts-logits-general · gemini-2.0-flash

    health waste DNA alarm development grievance seeding

    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
    SupportActionBar
    -0.49
     CharSequence
    -0.49
    magical
    -0.45
     anſ
    -0.43
     channels
    -0.42
    IBarButtonItem
    -0.42
     adolescents
    -0.41
    umque
    -0.40
    WithString
    -0.40
     istore
    -0.40
    POSITIVE LOGITS
    expandindo
    0.63
     Italijani
    0.61
    transQ
    0.59
    httphttps
    0.57
     disambiguazione
    0.54
     Италијани
    0.54
     ErrIntOverflow
    0.52
     ujednoznacz
    0.52
    انيف
    0.51
    UserScript
    0.48
    Activations Density 1.610%

    No Known Activations