© 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 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. 4827
    Prev
    Next
    INDEX
    Explanations

    terms related to geographical locations and organizations involved in conflicts

    oai_token-act-pair · gpt-3.5-turbo

    the term "al-Qaeda" and variations of it

    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
     AVG
    -0.79
    wcs
    -0.79
     Cosponsors
    -0.73
    OPLE
    -0.69
    ulhu
    -0.68
     Edited
    -0.68
    ongyang
    -0.67
     Dickinson
    -0.66
     Morty
    -0.65
    DragonMagazine
    -0.64
    POSITIVE LOGITS
    tra
    1.06
    Qaeda
    1.02
    Qaida
    0.98
    affiliated
    0.98
    bay
    0.96
    shaped
    0.95
    su
    0.94
    armed
    0.91
    associated
    0.90
    backed
    0.89
    Activations Density 0.100%

    No Known Activations