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. 8312
    Prev
    Next
    INDEX
    Explanations

    the use of constant and variable declarations in code

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

    variable or user declarations

    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
    Personendaten
    -0.69
    +#+#
    -0.66
    tvguidetime
    -0.64
    ftagPool
    -0.64
     ब्रेकडाउन
    -0.59
     désolés
    -0.59
    EDEFAULT
    -0.58
    حوالہ
    -0.57
     iNdEx
    -0.57
    twimg
    -0.55
    POSITIVE LOGITS
    import
    0.63
     K
    0.48
    BeginInit
    0.46
     Rock
    0.46
    The
    0.45
    K
    0.45
     import
    0.43
     John
    0.43
    Add
    0.43
     The
    0.42
    Activations Density 0.001%

    No Known Activations