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. 10-GEMMASCOPE-RES-16K
    6. 13200
    Prev
    Next
    INDEX
    Explanations

    metrics and data related to project evaluations and outcomes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     المعيارى
    -0.83
    MLLoader
    -0.73
    ulement
    -0.66
     الدولى
    -0.64
     Situs
    -0.62
    Replies
    -0.58
    KEYCODE
    -0.58
    SOUNDBITE
    -0.58
    itename
    -0.58
    posób
    -0.58
    POSITIVE LOGITS
    ISupport
    0.74
    awtextra
    0.57
     ostavi
    0.57
     дописавши
    0.54
    AndEndTag
    0.54
     unknownFields
    0.51
    ())));
    0.49
    In
    0.49
    WithIOException
    0.48
    onOptions
    0.47
    Activations Density 0.289%

    No Known Activations