© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 13886
    Prev
    Next
    INDEX
    Explanations

    negative impacts or disruptions caused by individuals or circumstances in various contexts

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

    unpleasant experiences

    np_acts-logits-general · gemini-2.0-flash

    trouble, pain, discomfort, terror

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     quedado
    -0.38
    colgante
    -0.30
     ramas
    -0.29
    muerzo
    -0.29
     coraggio
    -0.28
     queda
    -0.28
     değiş
    -0.28
     tewas
    -0.28
     colher
    -0.28
     quedan
    -0.28
    POSITIVE LOGITS
     Trouble
    0.74
     trouble
    0.73
    Trouble
    0.69
    trouble
    0.66
     disturbance
    0.64
     occaf
    0.63
     referenties
    0.62
    ergies
    0.62
    迷惑
    0.62
     troubles
    0.60
    Activations Density 0.269%

    No Known Activations