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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 15113
    Prev
    Next
    INDEX
    Explanations

    that one's really hard. I'll try my best.This neuron seems to activate for a variety of different sentence structures including being in or on, and also activates frequently with the word line.

    oai_token-act-pair · gemini-2.0-flash

    tokens related to technical or programming concepts

    np_token-act-pair-logits · gpt-4o-mini
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Samaria
    -0.61
    "]));
    -0.52
    ^(@)
    -0.52
    Freitag
    -0.50
    }")]
    -0.50
     breadth
    -0.49
     pò
    -0.49
     "}";
    -0.49
     }]
    -0.49
    )}_
    -0.48
    POSITIVE LOGITS
    ⓧ
    0.77
    GraphicsUnit
    0.71
    abetes
    0.67
     EconPapers
    0.61
     CreateTagHelper
    0.61
     pinulongan
    0.60
    Tembelea
    0.60
    ArgsConstructor
    0.57
    AnchorStyles
    0.57
     MenuView
    0.56
    Activations Density 1.443%

    No Known Activations