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

    Neuronpedia

    Jacobian LensNEW
    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. 20-GEMMASCOPE-RES-131K
    11. 71479
    Prev
    Next
    INDEX
    Explanations

    specific relationships and interactions within complex systems

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

    descriptive words followed by details

    np_acts-logits-general · gemini-2.5-flash-lite

    It detects the start-of-sequence (beginning of a document/turn).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
     
    -0.60
    ’
    -0.46
     The
    -0.41
    '
    -0.39
    begin
    -0.39
     the
    -0.37
     They
    -0.36
     A
    -0.36
      
    -0.36
    ↵
    -0.36
    POSITIVE LOGITS
     betweenstory
    1.37
    transQ
    1.18
    fromnode
    1.05
    RegressionTest
    1.00
     Numerade
    0.99
     queſta
    0.97
     Normdatei
    0.91
     Taktlose
    0.90
    <unused16>
    0.89
    <unused41>
    0.89
    Activations Density 7.009%

    No Known Activations