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-IT
    4. Residual Stream - 16k
    5. 9-GEMMASCOPE-RES-16K
    6. 6904
    Prev
    Next
    INDEX
    Explanations

    attributes indicating settings or states, particularly related to boolean values and visibility

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

    Boolean variable name prefixes

    np_acts-logits-general · gemini-2.0-flash

    has, is, have

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ArgsConstructor
    -0.81
     HasFactory
    -0.68
    MLLoader
    -0.67
    awtextra
    -0.67
    AndEndTag
    -0.65
     ब्रेकडाउन
    -0.64
    bootstrapcdn
    -0.62
     متعلقه
    -0.61
    новниш
    -0.59
    ScopeManager
    -0.58
    POSITIVE LOGITS
    has
    0.57
    Has
    0.54
     Has
    0.52
    OGND
    0.46
    Allow
    0.46
    show
    0.44
    skip
    0.40
    getIs
    0.40
    auto
    0.39
    have
    0.39
    Activations Density 0.548%

    No Known Activations