© 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. 18-GEMMASCOPE-TRANSCODER-16K
    7. 7746
    Prev
    Next
    INDEX
    Explanations

    the various conjugations of the verb 'be' ('is', 'are', 'being', 'been').

    oai_token-act-pair · gemini-2.0-flash

    being

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_18/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.18.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
    ########.
    -1.16
    Datuak
    -0.98
    +#+
    -0.94
    AddTagHelper
    -0.89
     ſtate
    -0.83
     houſe
    -0.82
    ">//
    -0.81
    TagMode
    -0.81
    IVEREF
    -0.80
     becauſe
    -0.80
    POSITIVE LOGITS
     try
    0.44
     get
    0.40
    try
    0.38
     build
    0.36
     look
    0.35
     verme
    0.35
     &
    0.34
     v
    0.33
     ins
    0.33
     so
    0.32
    Activations Density 3.294%

    No Known Activations