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. 784
    Prev
    Next
    INDEX
    Explanations

    code-related syntax and structures commonly used in programming

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

    DNA polymerase and related concepts

    np_acts-logits-general · gemini-2.0-flash

    polymerase, nucleotides, synthesis, assay, analogues

    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
     ModelExpression
    -0.73
    سكرية
    -0.43
    ⓧ
    -0.43
    jande
    -0.42
    jago
    -0.41
     acceptez
    -0.40
     Jensen
    -0.40
    iley
    -0.39
     Ghar
    -0.39
     отно
    -0.39
    POSITIVE LOGITS
     StringBuilder
    0.81
     polymerase
    0.76
     nucleotides
    0.74
    脚注の使い方
    0.71
     StringBuffer
    0.71
     append
    0.68
     Append
    0.68
     initComponents
    0.68
    LogFactory
    0.61
     Sonya
    0.60
    Activations Density 0.346%

    No Known Activations