© 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. 2-RES-JB
    11. 9213
    Prev
    Next
    INDEX
    Explanations

    adjectives and nouns commonly associated with the genre of superhero movies

    oai_token-act-pair · gpt-3.5-turbo

    references to superheroes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    choes
    -0.81
    pring
    -0.72
    imentary
    -0.72
    chard
    -0.72
    sweet
    -0.70
    ¿½
    -0.69
    enf
    -0.68
     Dise
    -0.68
    dated
    -0.67
    UD
    -0.67
    POSITIVE LOGITS
     superhero
    1.04
     superheroes
    0.94
     Comics
    0.80
     villain
    0.76
     Superman
    0.76
     comics
    0.75
    ically
    0.74
     heroes
    0.74
     mascot
    0.73
     costumes
    0.73
    Activations Density 0.016%

    No Known Activations