© 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 - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 31827
    Prev
    Next
    INDEX
    Explanations

    terms and phrases that indicate inadequacy or poor quality

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

    poor quality or execution

    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
     wyma
    -0.37
     Required
    -0.35
     registered
    -0.34
    required
    -0.33
     farms
    -0.33
     required
    -0.33
     çiçek
    -0.33
    Required
    -0.32
     turn
    -0.32
     registers
    -0.32
    POSITIVE LOGITS
     poorly
    0.88
     shitty
    0.85
     schlechte
    0.82
     slecht
    0.82
     crappy
    0.80
     disambiguazione
    0.77
     schlecht
    0.76
     inefficient
    0.73
     dårlig
    0.72
    nadequate
    0.71
    Activations Density 0.050%

    No Known Activations