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. 20-GEMMASCOPE-RES-16K
    6. 1431
    Prev
    Next
    INDEX
    Explanations

    phrases related to planning and scheduling activities or tasks

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

    preceding infinitives ("to")

    np_acts-logits-general · gemini-2.0-flash

    spending or time devotion

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    expandindo
    -0.92
     Roskov
    -0.55
    Instru
    -0.51
    offsetof
    -0.49
     ajud
    -0.47
    پیوند
    -0.47
    help
    -0.47
    gazione
    -0.46
    hilfe
    -0.45
     chính
    -0.45
    POSITIVE LOGITS
     spending
    0.73
    })*/
    0.72
     spend
    0.72
     melakukannya
    0.72
     spends
    0.67
     délais
    0.64
    جراء
    0.63
     devoting
    0.62
     Spending
    0.61
     spender
    0.61
    Activations Density 0.407%

    No Known Activations