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-2-2B
    4. Residual Stream - 65k
    5. 17-GEMMASCOPE-RES-65K
    6. 28462
    Prev
    Next
    INDEX
    Explanations

    occurrences of the name "Steve."

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    ный
    -0.71
    ת
    -0.71
    ing
    -0.67
    al
    -0.65
     Bort
    -0.64
     Tind
    -0.63
    alisa
    -0.59
     NOL
    -0.59
    —————
    -0.58
     Kunt
    -0.58
    POSITIVE LOGITS
    BeginContext
    0.92
     оригіналу
    0.92
    ioutil
    0.88
     Steve
    0.84
    httphttps
    0.84
     AssemblyCulture
    0.82
    STEVE
    0.82
    MemoryWarning
    0.81
     utafitiHapana
    0.80
     AppBundle
    0.73
    Activations Density 0.011%

    No Known Activations