© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 13393
    Prev
    Next
    INDEX
    Explanations

    code structure and function definitions in programming

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

    Code syntax and programming language elements

    np_acts-logits-general · gemini-2.0-flash

    code, math, logic, symbols

    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
     ffilmiau
    -0.66
    featureID
    -0.64
     noDo
    -0.63
     <<<<<<<<<<<<<<
    -0.60
    GOTREF
    -0.60
    |};
    -0.60
     betweenstory
    -0.58
    richTextPanel
    -0.58
     ContentValues
    -0.57
    hematics
    -0.57
    POSITIVE LOGITS
    jspb
    0.41
     terceira
    0.35
     étendue
    0.34
    PRNewswire
    0.33
     approche
    0.32
    toarele
    0.29
    itatem
    0.29
     abord
    0.29
    macht
    0.29
     Revolución
    0.28
    Activations Density 0.793%

    No Known Activations