© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 15150
    Prev
    Next
    INDEX
    Explanations

    constructions using the words "never", "ever" or "no"

    oai_token-act-pair · gemini-2.0-flash

    phrases indicating repetitiveness or negativity

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

    never/ever

    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
     always
    -2.08
    always
    -1.87
     Always
    -1.85
    Always
    -1.83
     sempre
    -1.82
     ever
    -1.80
     siempre
    -1.72
     never
    -1.68
     ALWAYS
    -1.63
    ALWAYS
    -1.59
    POSITIVE LOGITS
     AssemblyCulture
    0.59
    ielsen
    0.53
    +#+
    0.49
    بار
    0.46
     skjø
    0.45
    dfunding
    0.44
    labili
    0.44
    zurück
    0.43
    起来
    0.43
    TextAppearance
    0.42
    Activations Density 0.615%

    No Known Activations