© 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. 12768
    Prev
    Next
    INDEX
    Explanations

    scientific and technical words, especially words related to chemistry, biology and manufacturing

    oai_token-act-pair · gemini-2.0-flash

    DNA/RNA synthesis

    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
     produced
    -1.59
    made
    -1.55
     manufactured
    -1.53
    Made
    -1.52
    produced
    -1.51
     Produced
    -1.48
     made
    -1.45
     Made
    -1.41
     created
    -1.36
    manufactured
    -1.36
    POSITIVE LOGITS
     بخوانید
    0.48
    談社
    0.47
    сылкі
    0.46
     réessayer
    0.41
    ModelAndView
    0.41
    Jereo
    0.40
    Personendaten
    0.40
    strophic
    0.40
    tetten
    0.40
    InitVars
    0.40
    Activations Density 1.856%

    No Known Activations