© 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. 1-RES-JB
    11. 3547
    Prev
    Next
    INDEX
    Explanations

    phrases or words that indicate something being disproportionately more or less than expected

    oai_token-act-pair · gpt-3.5-turbo

    discussions of disproportionate impacts or effects on various groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ince
    -0.77
    ht
    -0.74
    ired
    -0.72
    uring
    -0.72
    held
    -0.71
    icist
    -0.70
    PT
    -0.70
    adal
    -0.70
    love
    -0.70
    ain
    -0.69
    POSITIVE LOGITS
     disproportionately
    1.07
     disproportion
    1.04
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    0.83
     disadvant
    0.81
     impacts
    0.81
    ãĤ¼ãĤ¦ãĤ¹
    0.80
     disadvantages
    0.79
     proport
    0.78
     adolesc
    0.78
     shenan
    0.76
    Activations Density 0.014%

    No Known Activations