© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 3118
    Prev
    Next
    INDEX
    Explanations

    elements of coding syntax or structure, particularly in programming or script writing contexts

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

    "In" followed by "since", "install", "recent", "strengths", "particular"

    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
     يتيمه
    -1.12
    niſſe
    -1.09
    <unused41>
    -1.05
    [@BOS@]
    -1.05
    <unused3>
    -1.05
    <unused43>
    -1.04
    <unused47>
    -1.04
    <unused16>
    -1.04
    <pad>
    -1.04
    <unused14>
    -1.04
    POSITIVE LOGITS
    1
    0.41
    ↵↵
    0.40
    3
    0.38
    2
    0.38
    8
    0.36
    4
    0.35
    7
    0.35
    5
    0.34
    9
    0.33
    Step
    0.33
    Activations Density 0.338%

    No Known Activations