© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 31-GEMMASCOPE-RES-131K
    11. 11960
    Prev
    Next
    INDEX
    Explanations

    programming constructs related to method returns and static definitions

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

    return Completion or case assignment

    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
     juntas
    -0.36
     juntos
    -0.31
     whatever
    -0.30
    centralwidget
    -0.29
     fáciles
    -0.29
     cuanto
    -0.28
     invitaciones
    -0.27
     gober
    -0.27
    ļas
    -0.26
     bro
    -0.25
    POSITIVE LOGITS
    SharedDtor
    0.75
    Tembelea
    0.74
     utafitiHapana
    0.74
     Administrativna
    0.70
     noDo
    0.67
    يكب
    0.66
    AndEndTag
    0.63
     للمعارف
    0.62
    ImageContext
    0.61
    UpInside
    0.59
    Activations Density 0.001%

    No Known Activations