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. 8513
    Prev
    Next
    INDEX
    Explanations

    references to death and family relationships

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

    death or survival

    np_acts-logits-general · gemini-2.0-flash

    death and outcomes

    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
    principalColumn
    -0.65
     mitu
    -0.54
    tagHelperRunner
    -0.50
    uses
    -0.49
    ⓘ
    -0.49
     vlak
    -0.48
     lenker
    -0.47
    USING
    -0.47
    
    -0.47
    thouse
    -0.47
    POSITIVE LOGITS
     survived
    1.04
    survi
    1.01
     Survi
    0.97
     survivors
    0.97
     surviving
    0.91
     Survivors
    0.85
     survive
    0.85
    Surviving
    0.84
     survives
    0.84
     SURV
    0.83
    Activations Density 0.022%

    No Known Activations