© 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. 1-RES-JB
    11. 20542
    Prev
    Next
    INDEX
    Explanations

    words related to speed or quickness

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "quickly."

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
    pel
    -0.80
    artisan
    -0.75
    quality
    -0.74
    orno
    -0.71
    rights
    -0.69
    unal
    -0.67
    pheus
    -0.67
    obyl
    -0.66
     commandments
    -0.66
    anyahu
    -0.66
    POSITIVE LOGITS
    aneously
    0.92
     paced
    0.84
    entimes
    0.81
     appre
    0.76
     thereafter
    0.74
     dissip
    0.74
     ensued
    0.73
     transitioned
    0.72
     asleep
    0.72
     succession
    0.72
    Activations Density 0.012%

    No Known Activations