© 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 - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 10918
    Prev
    Next
    INDEX
    Explanations

    humorous or comedic elements in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     PCS
    -0.46
     CIM
    -0.42
     SFR
    -0.42
    rrggbb
    -0.40
     PMA
    -0.40
     Domini
    -0.38
    ClientRect
    -0.38
     Superhost
    -0.38
     DSS
    -0.38
     MBR
    -0.38
    POSITIVE LOGITS
     comical
    0.78
     joking
    0.74
     joked
    0.70
     joke
    0.70
     humorous
    0.68
     laughable
    0.68
     comedic
    0.68
     jokingly
    0.68
     broma
    0.68
     irony
    0.68
    Activations Density 0.204%

    No Known Activations