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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 11546
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to analysis and comparison within data or results

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

    Precedes "entire" or words meaning "entire"

    np_acts-logits-general · gemini-2.0-flash

    entire or whole

    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
    Przypisy
    -0.62
    TagHelper
    -0.58
    tvguidetime
    -0.58
     فريبيس
    -0.56
     bezeichneter
    -0.56
    SharedCtor
    -0.55
     للاسماء
    -0.55
    NavController
    -0.54
     Italijani
    -0.53
    нительно
    -0.52
    POSITIVE LOGITS
     entire
    3.32
     whole
    2.98
    whole
    2.84
    entire
    2.75
     Entire
    2.66
     ENTIRE
    2.51
     entirety
    2.41
    Entire
    2.36
    Whole
    2.36
     WHOLE
    2.32
    Activations Density 0.785%

    No Known Activations