© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 11975
    Prev
    Next
    INDEX
    Explanations

    terms related to drugs and substances

    oai_token-act-pair · gpt-3.5-turbo

    terms related to drugs and their regulation

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    TeX
    -0.74
    borgh
    -0.68
    Weather
    -0.62
     Door
    -0.59
     Consent
    -0.59
     Location
    -0.58
     Alert
    -0.58
     Designs
    -0.57
    GBT
    -0.57
     Cargo
    -0.56
    POSITIVE LOGITS
     itself
    0.78
     synonymous
    0.75
    killer
    0.70
     shorthand
    0.70
    wikipedia
    0.69
    fighter
    0.68
     ze
    0.68
     stigma
    0.68
     scourge
    0.67
     known
    0.66
    Activations Density 0.466%

    No Known Activations