Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 93643
    Prev
    Next
    INDEX
    Explanations

    elements related to technical specifications and components in circuitry

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

    circuit descriptor

    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
     snippetHide
    -0.73
    
    -0.69
     laſſen
    -0.68
     المعيارى
    -0.67
     MainAxisSize
    -0.67
    +:+
    -0.65
    citroën
    -0.64
    RTCF
    -0.62
    <unused41>
    -0.62
    [@BOS@]
    -0.62
    POSITIVE LOGITS
     ModelRenderer
    0.44
    énario
    0.32
    ])),
    0.29
    fjspx
    0.29
    Să
    0.28
    DoubleQuotes
    0.28
    Set
    0.27
    Pranala
    0.27
    Uso
    0.27
    })),
    0.26
    Activations Density 0.005%

    No Known Activations