© 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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 98707
    Prev
    Next
    INDEX
    Explanations

    elements related to coding syntax or identifiers

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

    macho man

    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
    }{*}{}
    -0.40
    LabelTagHelper
    -0.40
     متعلقه
    -0.40
     Krim
    -0.38
    zają
    -0.38
    ɵɵelementStart
    -0.38
    Brunswick
    -0.37
     FetchType
    -0.36
    xane
    -0.36
     Pag
    -0.36
    POSITIVE LOGITS
    Jereo
    0.66
    oa̍t
    0.64
    treeview
    0.53
     Roskov
    0.53
     Préférences
    0.51
    scriptcase
    0.50
    estacks
    0.49
    تقاوى
    0.49
     Sequelize
    0.47
     referenties
    0.46
    Activations Density 0.000%

    No Known Activations