Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 0-GEMMASCOPE-TRANSCODER-16K
    4. 7150
    Prev
    Next
    INDEX
    Explanations

    the string "rem" (likely short for "remember") and the word "this"

    oai_token-act-pair · gemini-2.0-flash

    rem

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.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
     myſelf
    -1.91
     itſelf
    -1.83
     Efq
    -1.80
     Monfieur
    -1.76
     purpoſe
    -1.73
     ſtate
    -1.72
     pleaſure
    -1.70
     Reſ
    -1.70
     Houſe
    -1.68
     himſelf
    -1.66
    POSITIVE LOGITS
    .
    1.20
    ,
    1.02
     (
    1.02
     and
    0.91
    :
    0.88
    /
    0.84
     M
    0.83
     I
    0.82
     D
    0.82
     H
    0.81
    Activations Density 0.463%

    No Known Activations