© 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. 20455
    Prev
    Next
    INDEX
    Explanations

    specific locations or geographical features, particularly ones related to lakes

    oai_token-act-pair · gpt-3.5-turbo

    references to specific geographic locations, particularly those related to lakes

    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
    agnar
    -0.85
    imental
    -0.71
    udeb
    -0.71
    untu
    -0.70
    isible
    -0.70
    natal
    -0.70
    endiary
    -0.70
    ENTS
    -0.68
    inventoryQuantity
    -0.66
    acist
    -0.65
    POSITIVE LOGITS
     Lakes
    1.21
     Erie
    0.97
     Basin
    0.90
     lakes
    0.89
     ponds
    0.87
     basin
    0.80
    weed
    0.77
    creen
    0.74
     Reserv
    0.74
    hire
    0.73
    Activations Density 0.008%

    No Known Activations