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. 14539
    Prev
    Next
    INDEX
    Explanations

    instances of emotional expressions and feelings

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

    First few letters of a word

    np_acts-logits-general · gemini-2.0-flash

    foreign language

    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
     Bikin
    -0.65
    Personendaten
    -0.61
     Chriftian
    -0.59
     Chriſt
    -0.59
     Majefty
    -0.58
     eſt
    -0.58
    IndentedString
    -0.56
    neſs
    -0.54
    tront
    -0.52
    IUrlHelper
    -0.52
    POSITIVE LOGITS
    帖最后由
    0.69
     ķ
    0.62
     ilmo
    0.59
     ویکی‌پدیا
    0.59
    Embeddable
    0.58
     anvä
    0.58
     Chwiliwch
    0.58
     culoare
    0.58
     föres
    0.57
     izvē
    0.57
    Activations Density 0.294%

    No Known Activations