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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 105578
    Prev
    Next
    INDEX
    Explanations

    coding syntax and special characters often used in programming languages

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     pleaſure
    -0.51
     évêque
    -0.50
     plufieurs
    -0.50
     prêtre
    -0.47
     raiſ
    -0.46
     ſhe
    -0.43
     himſelf
    -0.42
     ſon
    -0.42
     ſte
    -0.41
     itſelf
    -0.40
    POSITIVE LOGITS
    +#+#
    0.56
     Jameson
    0.55
     CURIAM
    0.55
     Leger
    0.54
     ARTS
    0.54
    存于互联网档案馆
    0.54
    ActionTypes
    0.53
    夏休み
    0.53
    RegistryLite
    0.53
    半年
    0.52
    Activations Density 0.009%

    No Known Activations