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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 4660
    Prev
    Next
    INDEX
    Explanations

    terms related to the Marvel Cinematic Universe

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of the universe

    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
    marg
    -0.74
    mong
    -0.64
    let
    -0.62
    cent
    -0.61
     whistle
    -0.61
    wh
    -0.60
     sten
    -0.60
     carp
    -0.60
     bordering
    -0.58
     foot
    -0.57
    POSITIVE LOGITS
     Universe
    4.18
     universe
    2.34
     universes
    1.75
     cosmos
    1.70
     galaxy
    1.48
     galaxies
    1.43
     Milky
    1.34
     Galaxy
    1.29
     UNIVERS
    1.28
     Univers
    1.23
    Activations Density 0.008%

    No Known Activations