© 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. 99879
    Prev
    Next
    INDEX
    Explanations

    frequent use of definite articles and conjunctions in sentences

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

    USB, end, test, hand

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects numeric tokens, especially decimal/floating-point numbers.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
    resizingMask
    -0.47
     defStyleAttr
    -0.44
    AddTagHelper
    -0.42
    
    -0.41
     tartalomajánló
    -0.40
    twimg
    -0.40
     Unterhaltung
    -0.40
    Hozzá
    -0.40
    WebpackPlugin
    -0.40
    insegna
    -0.39
    POSITIVE LOGITS
     PeEnEo
    0.49
    ParallelGroup
    0.49
     ivelany
    0.43
    ponenten
    0.42
     Italijanski
    0.41
    ificity
    0.41
    SequentialGroup
    0.38
    ponent
    0.38
    جغرافيا
    0.38
    注
    0.38
    Activations Density 0.046%

    No Known Activations