© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 16683
    Prev
    Next
    INDEX
    Explanations

    references to the superhero "Superman"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "Superman" and associated characters

    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
    nel
    -0.76
    mia
    -0.71
     ranks
    -0.66
     awaits
    -0.65
    iak
    -0.65
    nels
    -0.64
    fr
    -0.63
    ri
    -0.62
    Map
    -0.62
    Cook
    -0.62
    POSITIVE LOGITS
     Superman
    3.66
     Luthor
    2.04
     Batman
    1.99
     Krypt
    1.96
    Batman
    1.90
     Godzilla
    1.76
     superheroes
    1.74
     superhero
    1.62
     Deadpool
    1.53
     Lois
    1.52
    Activations Density 0.024%

    No Known Activations