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. 20-GEMMASCOPE-RES-65K
    6. 1265
    Prev
    Next
    INDEX
    Explanations

    references to passive observation

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

    instances of passive observation and inaction

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

    sit or set

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     nakalista
    -0.63
    DoubleQuotes
    -0.60
     تعدى
    -0.57
    mtable
    -0.57
    μβρίου
    -0.56
    XmlAccessorType
    -0.56
    IUrlHelper
    -0.53
    ÚBL
    -0.51
    istle
    -0.51
    wikk
    -0.50
    POSITIVE LOGITS
     sit
    1.49
     Sit
    1.39
    Sit
    1.35
    sit
    1.24
     sits
    1.08
     sitting
    1.08
     relax
    1.06
    sitting
    1.05
     passive
    0.98
     sentarse
    0.98
    Activations Density 0.131%

    No Known Activations