© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 31-GEMMASCOPE-RES-131K
    11. 6889
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate mathematical calculations or expressions

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

    mathematical expressions inside parentheses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     this
    -0.52
     цього
    -0.41
     này
    -0.40
     этого
    -0.36
    this
    -0.36
    这是
    -0.35
    ']=='
    -0.34
     этому
    -0.33
     цьому
    -0.33
    </td>
    -0.33
    POSITIVE LOGITS
     above
    0.82
     aforementioned
    0.81
    最初の
    0.79
    above
    0.76
     firſt
    0.74
     initial
    0.74
    上面的
    0.71
     iniziale
    0.71
     initiale
    0.71
     oben
    0.70
    Activations Density 0.111%

    No Known Activations