© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 10493
    Prev
    Next
    INDEX
    Explanations

    the neuron seems to focus on identifying instances of sudden actions, involvement, or immersion, mixed with mentions of people and products.

    oai_token-act-pair · gemini-2.0-flash

    the word "hard"

    np_token-act-pair-logits · gpt-4o-mini

    hard

    np_max-act-logits · gemini-2.0-flash
    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
    ✥
    -0.53
    xase
    -0.48
    بية
    -0.46
     minas
    -0.43
    şak
    -0.43
     Schäfer
    -0.41
     gì
    -0.40
     Wachs
    -0.40
    dataIndex
    -0.39
    BoxLayout
    -0.39
    POSITIVE LOGITS
     hard
    1.15
     head
    1.05
    Hard
    1.01
     Hard
    0.93
    hard
    0.92
     HARD
    0.92
     headlong
    0.89
    HARD
    0.87
    FormTagHelper
    0.83
     hardest
    0.81
    Activations Density 2.425%

    No Known Activations