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
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 7203
    Prev
    Next
    INDEX
    Explanations

    variables, method signatures, and other programming constructs linked to potential buffer overflow vulnerabilities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 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
    #+#
    -0.60
    الحياه
    -0.49
    ंदीखरीदारी
    -0.48
     wireType
    -0.42
    StoryboardSegue
    -0.42
    Personensuche
    -0.41
    WriteBarrier
    -0.40
    AccessorTable
    -0.40
    PerformLayout
    -0.40
    rrggbb
    -0.40
    POSITIVE LOGITS
    Carex
    0.48
    incen
    0.48
    Bombay
    0.47
     Packs
    0.46
     Bob
    0.46
    ujud
    0.46
    ésult
    0.45
     bobs
    0.45
     Beginners
    0.44
    ayn
    0.44
    Activations Density 0.805%

    No Known Activations