© 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 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. 13100
    Prev
    Next
    INDEX
    Explanations

    terms associated with physical disabilities or impairments

    oai_token-act-pair · gpt-3.5-turbo

    references to disabilities and physical challenges

    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
     raven
    -0.82
    atus
    -0.78
    acted
    -0.74
     synth
    -0.70
    gered
    -0.67
     Ridley
    -0.67
    URI
    -0.66
     Vale
    -0.66
     pudding
    -0.65
    Blade
    -0.65
    POSITIVE LOGITS
     handic
    3.89
     mull
    1.28
     souven
    1.17
     handc
    1.11
     barric
    1.03
     cripp
    1.00
    ogy
    0.98
    christ
    0.94
     Negro
    0.94
     chall
    0.93
    Activations Density 0.054%

    No Known Activations