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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 4-RES-JB
    11. 12340
    Prev
    Next
    INDEX
    Explanations

    words related to different countries and institutions

    oai_token-act-pair · gpt-3.5-turbo

    references to males or groups of people identified as men

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

    No Comments

    Negative Logits
    Dog
    -0.78
    VICE
    -0.73
    BILL
    -0.71
     Pwr
    -0.71
    RAY
    -0.67
    TOP
    -0.66
     DM
    -0.65
    Dick
    -0.64
    Berry
    -0.62
    ENC
    -0.61
    POSITIVE LOGITS
    opausal
    1.25
    uscript
    1.06
    stru
    0.98
    istan
    0.97
    gling
    0.96
    emen
    0.95
    endez
    0.94
    thren
    0.88
    ifest
    0.86
    volent
    0.86
    Activations Density 0.010%

    No Known Activations