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. 9165
    Prev
    Next
    INDEX
    Explanations

    mentions of vast or all-encompassing entities such as universes

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of a "universe" or "cosmos."

    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
    atel
    -0.74
    enced
    -0.69
    BIL
    -0.68
    ongh
    -0.67
     Aid
    -0.64
    let
    -0.63
    activ
    -0.62
     Peck
    -0.62
     Polk
    -0.62
    ering
    -0.62
    POSITIVE LOGITS
     universe
    3.64
     Universe
    2.77
     universes
    2.53
     cosmos
    2.32
     galaxy
    2.05
     galaxies
    1.68
     world
    1.48
     Milky
    1.44
     cosmic
    1.38
    iverse
    1.37
    Activations Density 0.009%

    No Known Activations