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 - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 6930
    Prev
    Next
    INDEX
    Explanations

    sections of code or formatting in programming-related text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Set
    -0.49
     or
    -0.49
    lieri
    -0.47
     hukum
    -0.47
     (
    -0.46
    ign
    -0.45
    照
    -0.45
     Umständen
    -0.45
     Or
    -0.45
     o
    -0.43
    POSITIVE LOGITS
    StoryboardSegue
    1.04
     الحره
    0.99
    expandindo
    0.93
    ंदीखरीदारी
    0.86
     виправивши
    0.86
    IUrlHelper
    0.84
    tvguidetime
    0.80
     surla
    0.78
     Audiodateien
    0.78
    VersionUID
    0.77
    Activations Density 0.098%

    No Known Activations