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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 18-GEMMASCOPE-RES-16K
    6. 14676
    Prev
    Next
    INDEX
    Explanations

    questions related to reasoning and seeking explanations

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

    "Does" followed by a question

    np_acts-logits-general · gemini-2.0-flash

    forms questions with "does"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Мексичка
    -0.80
     المعيارى
    -0.77
     незавершена
    -0.77
     betweenstory
    -0.76
    DoubleQuotes
    -0.76
    bewerken
    -0.73
     Himo
    -0.73
    ThroughAttribute
    -0.72
    MLLoader
    -0.71
    SharedCtor
    -0.71
    POSITIVE LOGITS
     the
    0.87
     it
    0.73
     you
    0.71
     this
    0.62
     a
    0.61
     our
    0.60
    essing
    0.58
     anyone
    0.57
    udo
    0.55
     Tal
    0.53
    Activations Density 0.117%

    No Known Activations