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. 16-GEMMASCOPE-RES-16K
    6. 15716
    Prev
    Next
    INDEX
    Explanations

    scientific terms and names, particularly in genetic and biochemical contexts

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

    "Gene" as part of "QuantiGene"

    np_acts-logits-general · gemini-2.0-flash

    places often followed by a descriptive noun

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    -
    -0.48
    ly
    -0.46
     Unito
    -0.44
    баты
    -0.44
     the
    -0.42
    何况
    -0.42
    ­
    -0.41
     
    -0.41
     tắt
    -0.40
    FirstResponder
    -0.39
    POSITIVE LOGITS
     defaultstate
    1.05
     tartalomajánló
    1.03
    Personensuche
    0.95
    ftagPool
    0.95
    تقاوى
    0.92
    matchCondition
    0.87
    nicity
    0.86
    iffion
    0.85
     Jefus
    0.85
     protoimpl
    0.85
    Activations Density 1.003%

    No Known Activations