© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 3687
    Prev
    Next
    INDEX
    Explanations

    sections and references in structured documents, such as reports or papers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/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.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     trovar
    -0.71
     préférence
    -0.68
     jScrollPane
    -0.67
     Roskov
    -0.67
     Wikispecies
    -0.67
     Италијани
    -0.66
     oprot
    -0.65
     internationaux
    -0.64
    AsUp
    -0.64
     simplicité
    -0.63
    POSITIVE LOGITS
    RunAsync
    0.76
    subsection
    0.71
    NameInMap
    0.68
     defaultstate
    0.66
    subsubsection
    0.62
     cont
    0.62
    section
    0.61
    Datuak
    0.58
    OGND
    0.57
    pu
    0.53
    Activations Density 0.081%

    No Known Activations