© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 13104
    Prev
    Next
    INDEX
    Explanations

    text sequences that are used for censoring or redacting offensive or inappropriate language

    oai_token-act-pair · gpt-3.5-turbo

    sequences of asterisks or similar patterns in the text

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
     decl
    -0.67
     nitrogen
    -0.62
     temporal
    -0.61
     neutrality
    -0.59
    quickShipAvailable
    -0.59
     reimb
    -0.59
     Geo
    -0.59
    ãĤ·ãĥ£
    -0.58
     accompanying
    -0.57
    ular
    -0.57
    POSITIVE LOGITS
    ***
    3.98
    ****
    3.27
    *****
    3.13
    **
    2.89
     ***
    2.63
    ********
    2.55
    ************
    2.30
     ****
    2.14
    ****************
    2.02
    ********************************
    2.00
    Activations Density 0.017%

    No Known Activations