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. 15-GEMMASCOPE-RES-16K
    6. 8617
    Prev
    Next
    INDEX
    Explanations

    punctuations and formatting marks within text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    expandindo
    -0.82
    ########.
    -0.71
     تضيفلها
    -0.71
    kuuta
    -0.67
     Mérimée
    -0.62
     المعيارى
    -0.61
    preventDefault
    -0.60
     defaultstate
    -0.58
    Rohy
    -0.58
    Skocz
    -0.58
    POSITIVE LOGITS
     The
    0.66
     kasarigan
    0.58
    onets
    0.56
    posedge
    0.54
    urier
    0.53
    rufe
    0.52
    $',
    0.51
    "}>
    0.51
    linge
    0.51
    mits
    0.50
    Activations Density 0.129%

    No Known Activations