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. 20-GEMMASCOPE-RES-16K
    6. 212
    Prev
    Next
    INDEX
    Explanations

    conditional statements in the code

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

    Code containing conditional "if" or "else"

    np_acts-logits-general · gemini-2.0-flash

    conditional logic and structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Exacts
    -0.59
    twimg
    -0.58
     تضيفلها
    -0.58
     correctes
    -0.57
    WaitGroup
    -0.54
    britanni
    -0.53
    ธอ
    -0.53
    SQLiteDatabase
    -0.51
     AppDelegate
    -0.50
    glDelete
    -0.50
    POSITIVE LOGITS
    rungsseite
    0.90
    elif
    0.84
    elseif
    0.75
     elif
    0.74
     ELSE
    0.67
     else
    0.67
     ModelExpression
    0.61
    elsif
    0.61
     مرئيه
    0.60
     переписи
    0.59
    Activations Density 0.022%

    No Known Activations