© 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. 14703
    Prev
    Next
    INDEX
    Explanations

    text passages describing experimental quantum physics and chemistry

    oai_token-act-pair · gemini-2.0-flash

    substrings starting with 'a', 't', 'h', or 'w'

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

    technical, mixed text

    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
     itſelf
    -1.47
     myſelf
    -1.42
     Majefty
    -1.34
     faſt
    -1.32
     purpoſe
    -1.31
     pleaſure
    -1.31
     ſtate
    -1.30
     houſe
    -1.30
     ་་
    -1.27
     themſelves
    -1.26
    POSITIVE LOGITS
     ab
    0.75
     th
    0.70
     w
    0.66
     ha
    0.66
     t
    0.65
     ag
    0.63
     ro
    0.63
     dat
    0.62
     som
    0.62
     sal
    0.61
    Activations Density 3.238%

    No Known Activations