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. 20-GEMMASCOPE-TRANSCODER-16K
    4. 14063
    Prev
    Next
    INDEX
    Explanations

    past tense verbs following the word "that", and speculative statements about becoming rich

    oai_token-act-pair · gemini-2.0-flash

    becoming rich

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope/layer_20/width_16k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uxxxx
    -0.69
     CreateTagHelper
    -0.69
     wireType
    -0.68
     myſelf
    -0.66
    RoutingModule
    -0.66
    oa̍t
    -0.65
    ècie
    -0.65
    ibol
    -0.64
    ValueStyle
    -0.64
    bootstrapcdn
    -0.64
    POSITIVE LOGITS
     rich
    0.56
     become
    0.49
     richer
    0.48
     becomes
    0.47
    become
    0.46
     wealthy
    0.45
     proud
    0.42
     Become
    0.42
     became
    0.42
     becoming
    0.42
    Activations Density 0.966%

    No Known Activations