© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 9632
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and separators in a document

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

    Text in languages other than English

    np_acts-logits-general · gemini-2.0-flash

    Chinese punctuation and text

    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
    ,
    -0.96
     ly
    -0.59
     the
    -0.58
     on
    -0.58
     of
    -0.57
     no
    -0.57
     one
    -0.55
     -
    -0.54
     in
    -0.54
     '
    -0.54
    POSITIVE LOGITS
     فريبيس
    1.19
     itſelf
    1.15
    出版年
    1.00
    principalColumn
    0.99
     myſelf
    0.96
    tagHelperRunner
    0.94
     Shakspeare
    0.91
     pleaſure
    0.90
     مرئيه
    0.88
     سكانية
    0.87
    Activations Density 0.055%

    No Known Activations