© 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. 4280
    Prev
    Next
    INDEX
    Explanations

    This neuron seems to be firing on tokens (place, time, names, clubs) from several very different articles, and is likely just a meaningless match

    oai_token-act-pair · gemini-2.0-flash

    prepositions and locational phrases

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

    at

    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
     calendriers
    -0.75
    ]")]
    -0.69
     Vikipedi
    -0.68
    SequentialGroup
    -0.66
    MethodManager
    -0.65
    RegressionTest
    -0.63
    #+#
    -0.63
     siyang
    -0.63
     ModelExpression
    -0.63
    })*/
    -0.60
    POSITIVE LOGITS
     at
    2.58
     At
    1.49
    At
    1.42
    at
    1.39
     tại
    1.23
     AT
    0.90
     	e
    0.87
     ở
    0.85
     @
    0.82
    @
    0.78
    Activations Density 4.605%

    No Known Activations