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. 14552
    Prev
    Next
    INDEX
    Explanations

    complex mathematical expressions or structures

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

    LaTeX formatting

    np_acts-logits-general · gemini-2.0-flash

    LaTeX labels

    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
     referenties
    -0.84
    IsContent
    -0.80
     Efq
    -0.80
    }{*}{}
    -0.79
     Jefus
    -0.73
    ruptedException
    -0.72
     tiegħek
    -0.70
     importado
    -0.70
     vectorielle
    -0.69
     élector
    -0.69
    POSITIVE LOGITS
    脚注の使い方
    0.57
     &___
    0.51
    Kaynakça
    0.49
    igshid
    0.48
     BoxFit
    0.47
    ViewData
    0.47
     nahilalakip
    0.47
    :✨
    0.47
     butterknife
    0.45
     arşivlendi
    0.45
    Activations Density 0.154%

    No Known Activations