© 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. 9057
    Prev
    Next
    INDEX
    Explanations

    numerical values and their associated formats, particularly in parentheses or brackets

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

    Values inside parentheses

    np_acts-logits-general · gemini-2.0-flash

    accounting

    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
     kasarigan
    -0.75
     EconPapers
    -0.74
    WebVitals
    -0.69
    ViewFeatures
    -0.67
    日閲覧
    -0.65
    UrlResolution
    -0.63
    extAlignment
    -0.61
     CURLOPT
    -0.59
    Hentet
    -0.59
    Composable
    -0.59
    POSITIVE LOGITS
     equivalent
    0.76
    equivalent
    0.73
     equivalente
    0.67
     Equivalent
    0.64
     
    0.60
     roughly
    0.59
     approximately
    0.58
    oredCriteria
    0.57
     équi
    0.56
     equivale
    0.54
    Activations Density 0.263%

    No Known Activations