© 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. 0-GEMMASCOPE-RES-65K
    6. 9880
    Prev
    Next
    INDEX
    Explanations

    terms related to time, particularly referencing midnight and hydraulic systems

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.0.hook_resid_post
    Hook Layer
    0
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     مرئيه
    -0.60
     Wikimedijinoj
    -0.59
    <?
    -0.55
    enosis
    -0.54
     couche
    -0.54
    ctp
    -0.52
    western
    -0.52
    IBinder
    -0.52
    writerow
    -0.51
    łec
    -0.51
    POSITIVE LOGITS
     Celebr
    0.79
     celebr
    0.78
     ModelExpression
    0.67
    StoreMessageInfo
    0.62
    panic
    0.60
     panic
    0.60
    bindParam
    0.60
    Component
    0.60
     celebrated
    0.59
     сто
    0.59
    Activations Density 0.045%

    No Known Activations