Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 1421
    Prev
    Next
    INDEX
    Explanations

    technical specifications and features of devices

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

    Comes after a period or comma

    np_acts-logits-general · gemini-2.0-flash

    fee-free banking and technical features

    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
    Приятного
    -0.68
    ientôt
    -0.67
     jsonify
    -0.66
     "<?
    -0.65
    出版年
    -0.64
     evermore
    -0.61
    お楽しみに
    -0.60
     дописавши
    -0.58
    intenant
    -0.58
    ImageContext
    -0.58
    POSITIVE LOGITS
     nicely
    0.90
     surprisingly
    0.87
     impressively
    0.82
    Surprisingly
    0.79
     decently
    0.77
     Surprisingly
    0.75
     pleasantly
    0.72
     decent
    0.72
     thankfully
    0.72
     impressive
    0.68
    Activations Density 0.233%

    No Known Activations