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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 1900
    Prev
    Next
    INDEX
    Explanations

    instances of the word "control" and its variations in a programming context

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

    form and UI controls

    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
     Berlín
    -0.41
     unjustly
    -0.40
     Milliarden
    -0.40
     Administração
    -0.39
     millón
    -0.39
     Meksika
    -0.39
     importación
    -0.38
     Nasional
    -0.38
     Arxiv
    -0.38
    +#+#
    -0.38
    POSITIVE LOGITS
     Controls
    1.09
    controls
    1.05
    Controls
    1.05
    control
    0.93
     CONTROLS
    0.90
    Control
    0.88
     controls
    0.87
     Control
    0.85
     FormControl
    0.84
    UIControl
    0.80
    Activations Density 0.010%

    No Known Activations