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. 18-GEMMASCOPE-RES-16K
    6. 3077
    Prev
    Next
    INDEX
    Explanations

    references to the name "Kar" and its variations in different contexts

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

    starting with "Kar" or "kar"

    np_acts-logits-general · gemini-2.0-flash

    names starting with Kar

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     OnInit
    -0.60
     CommonModule
    -0.48
    rippa
    -0.47
    HttpPut
    -0.47
    omoto
    -0.47
    臣
    -0.46
     escu
    -0.46
     sn
    -0.45
    FOLLOW
    -0.45
    Holt
    -0.45
    POSITIVE LOGITS
     للمعارف
    0.88
    GEBURTSDATUM
    0.79
    enterOuterAlt
    0.79
    Datuak
    0.74
     Kar
    0.74
    Lähteet
    0.71
    postsleuth
    0.70
    Kar
    0.70
     invokingState
    0.69
    DllImport
    0.67
    Activations Density 0.060%

    No Known Activations