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. 18-GEMMASCOPE-RES-16K
    6. 3578
    Prev
    Next
    INDEX
    Explanations

    changes and variations in quantitative measurements and their relationships over time or conditions

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

    Changes or variations in a function

    np_acts-logits-general · gemini-2.0-flash

    change over time or across indices

    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
    endsection
    -0.64
     inmediato
    -0.59
    fromnode
    -0.58
     textStatus
    -0.57
    reszcie
    -0.56
    UrlEncoded
    -0.55
     définiti
    -0.55
    发表于
    -0.54
     disponibilités
    -0.54
    قایناقلار
    -0.54
    POSITIVE LOGITS
     throughout
    1.05
     across
    0.90
    throughout
    0.85
    隨著
    0.82
     overtime
    0.81
    随着
    0.81
     evolution
    0.80
     Laufe
    0.78
     evolve
    0.77
     evolves
    0.74
    Activations Density 0.929%

    No Known Activations