© 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. 3095
    Prev
    Next
    INDEX
    Explanations

    past tense verbs, especially when describing trends ("increased", "grown")

    oai_token-act-pair · gemini-2.0-flash

    variations of the word "fold"

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

    fold

    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
    __':
    -0.65
    ttino
    -0.60
    __":
    -0.59
    expandindo
    -0.58
    Démographie
    -0.57
     saites
    -0.57
    __":
    -0.55
    ]();
    -0.54
    withIdentifier
    -0.54
     Yarm
    -0.54
    POSITIVE LOGITS
    fold
    0.90
     fold
    0.84
    Fold
    0.71
     onAnimation
    0.69
     Fold
    0.69
    folds
    0.67
     folds
    0.65
    FOLD
    0.60
    folded
    0.56
    folding
    0.51
    Activations Density 2.167%

    No Known Activations