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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 21328
    Prev
    Next
    INDEX
    Explanations

    specific mentions of the term "Generation" with a focus on numbers 9 and 10

    oai_token-act-pair · gpt-3.5-turbo

    references to different "Generations," likely pertaining to demographics or technology generations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     grounds
    -0.73
     rub
    -0.70
    wine
    -0.69
     parcel
    -0.67
     carp
    -0.67
     flat
    -0.64
     ale
    -0.62
    wic
    -0.62
     abs
    -0.61
     flask
    -0.61
    POSITIVE LOGITS
     Generation
    4.26
    Gener
    1.89
     generation
    1.80
     Generations
    1.65
    generation
    1.55
     Gener
    1.27
     Millennials
    1.21
     Gen
    1.20
     GEN
    1.16
     generations
    1.16
    Activations Density 0.016%

    No Known Activations