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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 92670
    Prev
    Next
    INDEX
    Explanations

    references to physical discomfort or pain symptoms

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

    discomfort or worsening symptoms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ましょう
    -0.52
    だからこそ
    -0.51
    ならでは
    -0.48
    tranno
    -0.47
    だって
    -0.46
     помогут
    -0.46
    ArgumentParser
    -0.44
     yourself
    -0.44
     당신
    -0.44
     humanas
    -0.43
    POSITIVE LOGITS
     noticeably
    0.66
     intermittently
    0.65
     faintly
    0.61
     occasionally
    0.60
     almost
    0.59
     VERY
    0.57
    almost
    0.57
     everytime
    0.56
    Occasionally
    0.55
     occasional
    0.55
    Activations Density 0.108%

    No Known Activations