© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 14937
    Prev
    Next
    INDEX
    Explanations

    words that are positive in tone and sometimes appear within the phrase “a [positive word] thing”.

    oai_token-act-pair · gemini-2.0-flash

    good

    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
     best
    -0.77
    best
    -0.76
    Best
    -0.71
     Луч
    -0.68
     Best
    -0.64
     terbaik
    -0.61
    mejor
    -0.60
     meilleurs
    -0.58
    Луч
    -0.58
     beste
    -0.57
    POSITIVE LOGITS
     good
    1.51
    good
    1.28
    Good
    1.19
    GOOD
    1.10
     Good
    1.08
     GOOD
    1.05
     gud
    0.91
     buena
    0.82
    goods
    0.82
     goods
    0.81
    Activations Density 1.207%

    No Known Activations