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. 17-GEMMASCOPE-RES-16K
    6. 10789
    Prev
    Next
    INDEX
    Explanations

    names of geographical locations and rivers, particularly in Michigan

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/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.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ệc
    -0.57
     continúas
    -0.56
     terce
    -0.53
     perla
    -0.52
     rural
    -0.52
     Dons
    -0.51
     Zola
    -0.50
     vold
    -0.50
    laye
    -0.49
    __*/
    -0.48
    POSITIVE LOGITS
    tvguidetime
    0.59
    omock
    0.58
    suck
    0.53
     oreilles
    0.53
     indian
    0.53
    danno
    0.52
    omalainen
    0.52
    UMBIA
    0.52
    psico
    0.52
    aconda
    0.50
    Activations Density 0.092%

    No Known Activations