© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 6665
    Prev
    Next
    INDEX
    Explanations

    the auxiliary verbs "have," "has," "had," and contractions that include them, particularly when used with "you"

    oai_token-act-pair · gemini-2.0-flash

    Past-tense verbs

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
     been
    -1.82
    been
    -1.59
     BEEN
    -1.37
     Been
    -1.30
    Been
    -1.27
     ollut
    -0.96
     été
    -0.81
     vært
    -0.70
     بوده
    -0.65
     heen
    -0.63
    POSITIVE LOGITS
     blew
    0.74
     froze
    0.68
     took
    0.68
     broke
    0.66
     flew
    0.65
     tore
    0.65
     gave
    0.65
     went
    0.65
     drew
    0.63
     returned
    0.62
    Activations Density 0.708%

    No Known Activations