© 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. 3-RES-JB
    11. 16596
    Prev
    Next
    INDEX
    Explanations

    names related to an individual named "Yoshi" and terms related to a controversy involving "karoshi" (death by overwork)

    oai_token-act-pair · gpt-3.5-turbo

    names of individuals associated with significant cultural or gaming contributions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    getic
    -0.75
    lain
    -0.75
    oline
    -0.71
    ffield
    -0.70
     Gazette
    -0.64
    ansas
    -0.64
    eworld
    -0.64
    ophy
    -0.63
     opener
    -0.63
    gling
    -0.63
    POSITIVE LOGITS
    oshi
    0.97
    imura
    0.96
    amoto
    0.94
    azaki
    0.91
    pei
    0.91
     Yosh
    0.91
     Hiro
    0.90
    ya
    0.87
     Yen
    0.87
    aki
    0.86
    Activations Density 0.018%

    No Known Activations