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. 20-GEMMASCOPE-RES-16K
    6. 16277
    Prev
    Next
    INDEX
    Explanations

    coding structures and function definitions

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

    "Res", "resolution", or "residue"

    np_acts-logits-general · gemini-2.0-flash

    resolution or resource identifier

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     utafitiHapana
    -0.57
     knitting
    -0.49
    CreateInfo
    -0.47
    aita
    -0.46
     cár
    -0.45
     Guanajuato
    -0.44
     officielles
    -0.43
     alugar
    -0.42
     officier
    -0.42
     athe
    -0.42
    POSITIVE LOGITS
     Res
    2.35
    Res
    2.32
     res
    2.29
     RES
    2.26
    res
    2.22
    RES
    2.13
     resolution
    1.90
     Resolution
    1.72
     rés
    1.72
    Resolution
    1.67
    Activations Density 0.151%

    No Known Activations