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. 15-GEMMASCOPE-RES-16K
    6. 1454
    Prev
    Next
    INDEX
    Explanations

    conversations or exchanges involving emotional connections and farewells

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     FetchType
    -0.58
     estimés
    -0.53
    brido
    -0.49
    alysis
    -0.49
    SafeMath
    -0.47
    出版年
    -0.47
    =$?
    -0.47
    rozco
    -0.47
     للمعارف
    -0.47
    BagLayout
    -0.46
    POSITIVE LOGITS
     goodbye
    0.84
    leaving
    0.83
     leaving
    0.79
     leave
    0.75
     Goodbye
    0.73
    Exeunt
    0.72
     meninggalkan
    0.72
     departure
    0.70
    leave
    0.68
     parting
    0.67
    Activations Density 0.190%

    No Known Activations