© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 70789
    Prev
    Next
    INDEX
    Explanations

    emotional responses or tensions in stressful situations

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

    in shock, evacuations headed, realized there, precision this

    np_acts-logits-general · gemini-2.5-flash-lite

    the neuron detects words and phrases appearing in vivid first-person accounts of acute physical emergencies or intense bodily events (e.g., childbirth, injuries, medical crises).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
    parsedMessage
    -0.54
    Демографія
    -0.53
    forName
    -0.49
     today
    -0.47
     Interest
    -0.47
     gezet
    -0.47
     normative
    -0.46
    andaag
    -0.46
    Today
    -0.45
    Gizmos
    -0.45
    POSITIVE LOGITS
     myſelf
    0.51
     CreateTagHelper
    0.43
    fortawesome
    0.42
    }))
    0.36
     ſte
    0.36
    "]);
    0.35
     themſelves
    0.34
     nonUne
    0.34
    CROSSTALK
    0.33
     paſſ
    0.33
    Activations Density 0.032%

    No Known Activations