© 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. 8-RES-JB
    11. 13119
    Prev
    Next
    INDEX
    Explanations

    descriptions of things being easy or light

    oai_token-act-pair · gpt-3.5-turbo

    descriptors that indicate size, comfort, and weight of objects

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    åŃ
    -0.73
    ictions
    -0.72
    icist
    -0.69
    ussions
    -0.69
    pta
    -0.68
     Peninsula
    -0.67
    cknow
    -0.66
     è£ıç
    -0.65
    National
    -0.65
    xit
    -0.65
    POSITIVE LOGITS
     matte
    1.11
     sturdy
    1.10
     waterproof
    1.09
     removable
    1.04
     lightweight
    1.01
     aest
    1.01
     adjustable
    1.01
     erg
    1.00
     bulky
    0.99
     durable
    0.97
    Activations Density 0.714%

    No Known Activations