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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 21-GEMMASCOPE-TRANSCODER-16K
    4. 6084
    Prev
    Next
    INDEX
    Explanations

    the word "might" and related words like "could"

    oai_token-act-pair · gemini-2.0-flash

    the word "might"

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

    suggestions and possibilities

    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
     maybe
    -0.75
    maybe
    -0.71
     Maybe
    -0.71
    perhaps
    -0.71
    Maybe
    -0.70
     better
    -0.69
     perhaps
    -0.69
     probably
    -0.66
     Probably
    -0.65
    probably
    -0.65
    POSITIVE LOGITS
     оригіналу
    0.60
     surla
    0.56
     wants
    0.55
     not
    0.53
     want
    0.52
    StreetMap
    0.50
    toHaveBeenCalled
    0.49
     kiek
    0.49
    typeparam
    0.49
    anément
    0.48
    Activations Density 0.638%

    No Known Activations