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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 22-GEMMASCOPE-RES-16K
    6. 13090
    Prev
    Next
    INDEX
    Explanations

    technical terms related to engineering and machinery

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

    Comes after "employ" or "a"

    np_acts-logits-general · gemini-2.0-flash

    specific components and mechanisms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     CreateTagHelper
    -0.62
     lengkap
    -0.58
     pior
    -0.55
    onActivityResult
    -0.54
     peores
    -0.53
     endommag
    -0.52
     oídos
    -0.52
    Sklici
    -0.51
    شهاد
    -0.51
     caract
    -0.50
    POSITIVE LOGITS
     specially
    0.95
     special
    0.93
     спеціаль
    0.89
     specjal
    0.87
     speciali
    0.83
    Personensuche
    0.83
     specialized
    0.79
     speciale
    0.79
    specially
    0.78
    special
    0.76
    Activations Density 1.139%

    No Known Activations