© 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. 31-GEMMASCOPE-RES-131K
    6. 93275
    Prev
    Next
    INDEX
    Explanations

    instances of the word "read" and the mention of numerical data or steps

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

    counting down numbers

    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
     surla
    -0.82
     незавершена
    -0.69
    Tembelea
    -0.59
    UnitTesting
    -0.56
     HasFactory
    -0.53
     Normdatei
    -0.53
     '\\;'
    -0.53
    MessageOf
    -0.52
    RectangleBorder
    -0.52
    titleMargin
    -0.50
    POSITIVE LOGITS
     count
    0.88
     counting
    0.84
     countdown
    0.79
     COUNT
    0.77
    count
    0.74
     Count
    0.72
     counted
    0.71
    Count
    0.69
     counts
    0.69
    COUNT
    0.68
    Activations Density 0.042%

    No Known Activations