© 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 - 131k
    9. 31-GEMMASCOPE-RES-131K
    11. 110156
    Prev
    Next
    INDEX
    Explanations

    specific sequences and structures related to a mathematical or scientific context, particularly involving the notation of entities labeled as 'AB'

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

    AB followed by common delimiters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Dichtung
    -0.46
     nahilalakip
    -0.44
     Verteilung
    -0.43
     Lingkungan
    -0.43
     Empfang
    -0.43
     McCa
    -0.41
     Occidente
    -0.41
    Память
    -0.40
     Ausbau
    -0.40
     gordura
    -0.40
    POSITIVE LOGITS
     AB
    0.67
    AB
    0.63
    BASELINE
    0.55
    setVerticalGroup
    0.54
     Wire
    0.48
    NAB
    0.48
     LAB
    0.47
    autorest
    0.46
     RAB
    0.46
    writeFieldEnd
    0.46
    Activations Density 0.017%

    No Known Activations