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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 13-GEMMASCOPE-RES-65K
    6. 39204
    Prev
    Next
    INDEX
    Explanations

    phrases related to turning things on or off

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.13.hook_resid_post
    Hook Layer
    13
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    thâu
    -0.74
    sidemargin
    -0.65
    IsMutable
    -0.65
     EconPapers
    -0.64
     resourceCulture
    -0.62
     newBuilder
    -0.62
    Puis
    -0.61
    RegistryLite
    -0.59
    ."));
    -0.59
    geist
    -0.58
    POSITIVE LOGITS
    openConnection
    0.96
     apag
    0.74
    glEnable
    0.74
     lights
    0.73
     Activ
    0.73
     encender
    0.71
    activate
    0.71
     OFF
    0.70
     OnEnable
    0.69
    Lights
    0.69
    Activations Density 0.018%

    No Known Activations