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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 105914
    Prev
    Next
    INDEX
    Explanations

    elements related to quantity, measurement, or programming functions

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

    decimal numbers and percentages

    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
    Sicherheit
    -0.50
     IFTT
    -0.47
     simplifié
    -0.46
    Practical
    -0.45
     masculinos
    -0.44
    Практи
    -0.44
     Fiscalía
    -0.43
    Хьажоргаш
    -0.42
    atschappij
    -0.42
    DllImport
    -0.42
    POSITIVE LOGITS
    tagHelperRunner
    0.53
    DockStyle
    0.51
     resourceCulture
    0.43
    AnchorTagHelper
    0.42
    OGND
    0.40
     '\\;'
    0.40
     intellectually
    0.40
    MLLoader
    0.40
     intellectual
    0.40
    onCreate
    0.40
    Activations Density 0.071%

    No Known Activations