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-2-9B-IT
    4. Residual Stream - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 20574
    Prev
    Next
    INDEX
    Explanations

    expressions related to amusement or entertainment

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

    amusement, Appellee, arousal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     vectorielle
    -0.57
     poitrine
    -0.55
     näin
    -0.53
     getOrder
    -0.52
     CascadeType
    -0.50
     HttpHeaders
    -0.50
     Linger
    -0.50
     Longo
    -0.50
    brtc
    -0.50
     vectorielles
    -0.50
    POSITIVE LOGITS
     amusement
    1.45
     amused
    1.34
     amuse
    1.30
     amusing
    1.19
     amusements
    1.08
     amus
    1.05
    usement
    1.02
     divertimento
    0.78
     diversión
    0.73
     divertido
    0.72
    Activations Density 0.005%

    No Known Activations