© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 15294
    Prev
    Next
    INDEX
    Explanations

    words related to people and doing, such as "who", "they", "to" and "could".

    oai_token-act-pair · gemini-2.0-flash

    pronouns and auxiliary verbs

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

    who

    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
    AccessorTable
    -0.76
    ьаж
    -0.72
     ddelweddau
    -0.71
    躇
    -0.71
    itinéraire
    -0.68
     дописавши
    -0.67
     NSCoder
    -0.66
     lts
    -0.65
    PasswordEncoder
    -0.63
    noinspection
    -0.63
    POSITIVE LOGITS
     we
    0.60
     am
    0.60
     knows
    0.60
     you
    0.58
     doesn
    0.57
     wants
    0.56
     knew
    0.54
     are
    0.52
     cannot
    0.51
     hasn
    0.50
    Activations Density 1.636%

    No Known Activations