© 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. 7425
    Prev
    Next
    INDEX
    Explanations

    instances where something is physically or figuratively close in proximity

    oai_token-act-pair · gpt-3.5-turbo

    phrases or contexts indicating proximity or becoming nearer to something

    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
    anity
    -0.78
    Ox
    -0.77
    EVA
    -0.73
    NZ
    -0.72
    hal
    -0.71
    DN
    -0.71
    SS
    -0.69
    plain
    -0.68
    DOM
    -0.68
    tech
    -0.68
    POSITIVE LOGITS
     closer
    1.08
     "$:/
    0.92
     examination
    0.90
     scrutiny
    0.89
     referen
    0.84
     confid
    0.82
     proximity
    0.82
     clos
    0.81
     nearer
    0.81
     dilig
    0.80
    Activations Density 0.007%

    No Known Activations