© 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. 1446
    Prev
    Next
    INDEX
    Explanations

    first-person pronouns combined with expressions of opinion or belief

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

    i guess, believe, fear

    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
    ंदीखरीदारी
    -0.77
    SharedDtor
    -0.64
    RTLI
    -0.59
     Biôgrafia
    -0.58
    Personendaten
    -0.57
     CanadaChoose
    -0.56
     Majefty
    -0.55
     itſelf
    -0.54
     fashiola
    -0.54
     myſelf
    -0.54
    POSITIVE LOGITS
    migrationBuilder
    0.36
     RoutedEventArgs
    0.35
     scoperto
    0.34
     imo
    0.32
    ResumeLayout
    0.32
     says
    0.32
     EventArgs
    0.31
     מוצ
    0.31
     true
    0.31
     said
    0.30
    Activations Density 0.026%

    No Known Activations