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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 997
    Prev
    Next
    INDEX
    Explanations

    terms associated with distribution models in statistics and mathematics

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

    "Negative" or "Illegal" followed by a noun

    np_acts-logits-general · gemini-2.0-flash

    invalid or negative inputs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
    foglio
    -0.41
    buttonBar
    -0.40
    Facility
    -0.35
    iastical
    -0.35
    ultados
    -0.35
     jeugd
    -0.35
     informée
    -0.35
    ใช่
    -0.35
     memanfaatkan
    -0.35
     canadensis
    -0.35
    POSITIVE LOGITS
    principalColumn
    0.60
    tagHelperRunner
    0.52
    velocityY
    0.51
    DIPSETTING
    0.48
     يتيمه
    0.48
     Chwiliwch
    0.48
    ftagPool
    0.47
     Infórmanos
    0.46
     estekak
    0.46
     betweenstory
    0.45
    Activations Density 0.686%

    No Known Activations