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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 68137
    Prev
    Next
    INDEX
    Explanations

    forms of the verb "to be" across various contexts

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

    past participles after 'is', 'was', 'were', 'did'

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     arşivlendi
    -0.39
    WillAppear
    -0.38
    forKey
    -0.37
    styleType
    -0.34
    انجليز
    -0.33
    Personendaten
    -0.32
    ParallelGroup
    -0.32
    Geplaatst
    -0.32
    GraphicsUnit
    -0.32
     hendes
    -0.31
    POSITIVE LOGITS
    RenderAtEndOf
    0.61
    الحياه
    0.58
    zty
    0.57
    LookAnd
    0.50
    PORTER
    0.50
     CreateTagHelper
    0.50
    downe
    0.49
    Documents
    0.48
    Consensus
    0.48
     ſeinen
    0.48
    Activations Density 0.178%

    No Known Activations