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. 18-GEMMASCOPE-RES-16K
    6. 1485
    Prev
    Next
    INDEX
    Explanations

    elements related to organization and communication in community meetings

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

    promotion or notification

    np_acts-logits-general · gemini-2.0-flash

    advertising and promotion

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    CloseOperation
    -0.66
     getItemId
    -0.63
    HomeAsUpEnabled
    -0.59
     OFDb
    -0.59
    RenderAtEndOf
    -0.58
     NSCoder
    -0.57
    RequiresApi
    -0.57
    rhosis
    -0.55
     оригіналу
    -0.55
    Uninitialized
    -0.54
    POSITIVE LOGITS
     publicity
    1.48
     publici
    1.45
     advertise
    1.37
     advertising
    1.32
     publicized
    1.31
     advertisement
    1.28
     promotion
    1.25
     advertised
    1.23
     promotional
    1.21
     dissemination
    1.18
    Activations Density 0.280%

    No Known Activations