© 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-2-2B
    4. Residual Stream - 65k
    5. 13-GEMMASCOPE-RES-65K
    6. 14900
    Prev
    Next
    INDEX
    Explanations

    instances of the word "when" and related conditional phrases indicating time or circumstances

    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
     Pelop
    -0.77
    PhysRev
    -0.77
    bewerken
    -0.76
    참고
    -0.76
     OGS
    -0.76
     useAppContext
    -0.72
     PDI
    -0.72
     ―――――
    -0.72
     OSA
    -0.71
     ་་
    -0.71
    POSITIVE LOGITS
     it
    0.71
     someone
    0.63
     
    0.57
     every
    0.57
     that
    0.55
     the
    0.55
     they
    0.54
     he
    0.54
     you
    0.52
     let
    0.50
    Activations Density 0.044%

    No Known Activations