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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 35540
    Prev
    Next
    INDEX
    Explanations

    descriptions of physical and emotional deterioration or hardship

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

    worn out or damaged

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    IntoConstraints
    -0.62
     surla
    -0.59
    EndProject
    -0.52
    Personendaten
    -0.52
    GOTREF
    -0.51
    ьаж
    -0.48
    Gambas
    -0.48
    SuccessListener
    -0.46
    Vidite
    -0.46
    تقاوى
    -0.45
    POSITIVE LOGITS
     damaged
    0.79
    Damaged
    0.73
     Damaged
    0.66
    damaged
    0.65
     wounded
    0.61
     broken
    0.61
     shattered
    0.59
     scarred
    0.59
     bescha
    0.59
     tired
    0.58
    Activations Density 0.050%

    No Known Activations