© 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. 15504
    Prev
    Next
    INDEX
    Explanations

    references to activities involving dancing

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "dancing."

    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
    chens
    -0.71
    LR
    -0.69
    Intel
    -0.69
    termin
    -0.69
     IP
    -0.67
    author
    -0.67
    Hig
    -0.65
    Pacific
    -0.65
    JM
    -0.65
     Tier
    -0.65
    POSITIVE LOGITS
     dancing
    3.53
     dancers
    2.78
     dances
    2.77
     danced
    2.68
     dancer
    2.57
     dance
    2.47
     Dancing
    2.46
     Dance
    1.95
     ballet
    1.76
     singing
    1.61
    Activations Density 0.023%

    No Known Activations