Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 12-GEMMASCOPE-RES-16K
    6. 9663
    Prev
    Next
    INDEX
    Explanations

    emotionally charged phrases relating to crowds and audience reactions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    PMailer
    -0.80
    enumii
    -0.60
     &___
    -0.59
    verifyException
    -0.58
    égalité
    -0.57
    TargetException
    -0.57
    tvguidetime
    -0.55
     dien
    -0.53
     viewType
    -0.53
    AccessorTable
    -0.52
    POSITIVE LOGITS
     cheering
    0.84
     crowd
    0.76
    crowd
    0.75
     spectators
    0.73
    Crowd
    0.71
     crowds
    0.69
     rooting
    0.68
     Crowd
    0.66
     fans
    0.65
     BoxFit
    0.62
    Activations Density 0.121%

    No Known Activations