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 - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 13842
    Prev
    Next
    INDEX
    Explanations

    references to loved ones and familial relationships

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

    "ones" after "loved" or "loved ones"

    np_acts-logits-general · gemini-2.0-flash

    loved ones or one

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    invokeLater
    -0.67
     समीक्षाओं
    -0.66
    SerializedSize
    -0.64
     actionMode
    -0.62
     Мексичка
    -0.59
    DateTimeField
    -0.56
    IsMutable
    -0.56
     Roskov
    -0.56
    setupUi
    -0.54
     Pristupljeno
    -0.53
    POSITIVE LOGITS
     ones
    1.85
     Ones
    1.66
    Ones
    1.40
     One
    1.37
    ones
    1.37
    One
    1.27
     ONE
    1.21
     one
    1.09
    one
    1.08
    ONE
    1.03
    Activations Density 0.205%

    No Known Activations