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. 2886
    Prev
    Next
    INDEX
    Explanations

    references to specific days and moments in time

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

    "[Time unit] before/after [something]"

    np_acts-logits-general · gemini-2.0-flash

    days, nights, and moments

    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
    ſelves
    -0.79
    FundMe
    -0.69
    uxxxx
    -0.64
    ynos
    -0.64
    StoreMessageInfo
    -0.63
     Perſ
    -0.63
     Савезне
    -0.62
    MessageState
    -0.62
     Reſ
    -0.62
     Houſe
    -0.61
    POSITIVE LOGITS
     they
    1.28
     we
    1.19
     that
    1.18
     you
    1.12
     it
    0.94
     someone
    0.89
     he
    0.87
     she
    0.79
     somebody
    0.73
     the
    0.65
    Activations Density 0.213%

    No Known Activations