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. 29586
    Prev
    Next
    INDEX
    Explanations

    function call syntax and structure in programming code

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

    code comments and punctuation

    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
    تقاوى
    -0.88
     resourceCulture
    -0.87
    ValueStyle
    -0.85
     tartalomajánló
    -0.84
    +#+#
    -0.83
     CreateTagHelper
    -0.79
    AddTagHelper
    -0.77
     pinulongan
    -0.77
     Taktlose
    -0.76
     يتيمه
    -0.75
    POSITIVE LOGITS
     témoin
    0.36
     maternidade
    0.34
     näky
    0.33
    太
    0.32
     aperçu
    0.31
    rą
    0.31
     helados
    0.31
     Jerusalén
    0.31
     electrónico
    0.30
     policiales
    0.30
    Activations Density 0.008%

    No Known Activations