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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 86546
    Prev
    Next
    INDEX
    Explanations

    phrases and keywords related to buffer overflow vulnerabilities in programming contexts

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

    `\x0a</data></write>`

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     biological
    -0.35
     atlik
    -0.34
    Departure
    -0.33
     tillgäng
    -0.33
     work
    -0.32
     natuurlijke
    -0.32
     bakgrund
    -0.31
     established
    -0.31
     agencies
    -0.29
     nouveautés
    -0.29
    POSITIVE LOGITS
    tagHelperRunner
    0.61
    CppMethod
    0.59
    endpush
    0.58
    verifyException
    0.57
     Meksiku
    0.57
    Искәрмәләр
    0.55
     OMITBAD
    0.53
     shootout
    0.52
     Chwiliwch
    0.52
     Roskov
    0.50
    Activations Density 0.015%

    No Known Activations