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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 8-RES-JB
    11. 12286
    Prev
    Next
    INDEX
    Explanations

    names of characters or superheroes in the X-Men universe

    oai_token-act-pair · gpt-3.5-turbo

    references to characters in comic book teams

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     formulations
    -0.90
    projects
    -0.85
    financial
    -0.84
    ourse
    -0.82
    ements
    -0.82
     sourcing
    -0.80
    POST
    -0.79
     publications
    -0.79
     filings
    -0.78
    conom
    -0.76
    POSITIVE LOGITS
     Vampire
    1.14
     Berserker
    1.04
     Minion
    1.04
     Werewolf
    1.04
     Fenrir
    1.03
     Tyrann
    1.02
     pengu
    1.01
     Luffy
    1.00
     Gladiator
    1.00
     Kenobi
    0.98
    Activations Density 0.266%

    No Known Activations