© 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. 58104
    Prev
    Next
    INDEX
    Explanations

    elements related to data structure representations, specifically nodes, edges, and their attributes in a graphical or network context

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

    attribute assignments or configuration keys

    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
    
    
    -0.46
     ExecuteAsync
    -0.40
    waitFor
    -0.40
    ፩
    -0.39
     Photocase
    -0.38
    ambién
    -0.37
     metropolitana
    -0.37
    Географиясе
    -0.35
    harris
    -0.35
    énier
    -0.35
    POSITIVE LOGITS
    twimg
    0.63
    GEBURTSDATUM
    0.55
     transfieras
    0.52
     AssemblyProduct
    0.48
     Biôgrafia
    0.48
     AssemblyTitle
    0.47
    хьтан
    0.47
    +#+#
    0.45
    stateMutability
    0.45
    ...".
    0.45
    Activations Density 0.163%

    No Known Activations