Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 103035
    Prev
    Next
    INDEX
    Explanations

    specific programming language functions or commands

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
     stay
    -0.48
     potion
    -0.47
     Sinks
    -0.46
    estival
    -0.46
     Thermostat
    -0.46
    δί
    -0.46
     viol
    -0.45
    Vesti
    -0.44
    SAT
    -0.44
     ipt
    -0.44
    POSITIVE LOGITS
    AddRange
    3.94
     propOrder
    0.87
    addAll
    0.82
     autorytatywna
    0.73
     initComponents
    0.72
    AddAttribute
    0.68
    MigrationBuilder
    0.67
    AddTagHelper
    0.66
    addImage
    0.66
     defaultstate
    0.66
    Activations Density 0.000%

    No Known Activations