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. 14356
    Prev
    Next
    INDEX
    Explanations

    topics related to health issues and injuries

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

    Part of something, especially negative events

    np_acts-logits-general · gemini-2.0-flash

    normal or unavoidable events

    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
     Reich
    -0.53
    utnik
    -0.44
     ta
    -0.44
    Reich
    -0.42
     co
    -0.41
    VERT
    -0.41
     Al
    -0.41
     ex
    -0.41
    idia
    -0.41
     idea
    -0.41
    POSITIVE LOGITS
     unavoidable
    1.29
     inevitable
    1.22
     hoort
    1.18
     normaux
    1.11
    inev
    1.06
     normais
    1.03
    GEBURTSDATUM
    1.02
     inev
    1.01
     inevit
    1.00
     naturais
    0.98
    Activations Density 0.370%

    No Known Activations