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. 1051
    Prev
    Next
    INDEX
    Explanations

    references to mathematical methods and techniques in research contexts

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

    Methods or techniques

    np_acts-logits-general · gemini-2.0-flash

    using standard techniques

    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
     propOrder
    -0.71
    PreExecute
    -0.69
    expandindo
    -0.67
    WriteBarrier
    -0.65
    IntoConstraints
    -0.64
    styleable
    -0.61
    ViewImports
    -0.61
    kelijkheid
    -0.61
     समीक्षक
    -0.60
     bParam
    -0.59
    POSITIVE LOGITS
     methods
    1.41
    methods
    1.34
     techniques
    1.32
     method
    1.31
     méthode
    1.24
     metode
    1.22
     métodos
    1.21
    method
    1.19
     Methods
    1.19
     méthodes
    1.18
    Activations Density 1.666%

    No Known Activations