© 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. 2-RES-JB
    11. 18503
    Prev
    Next
    INDEX
    Explanations

    negative or dangerous concepts, specifically related to evil or harm

    oai_token-act-pair · gpt-3.5-turbo

    references to malevolent or harmful entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    allas
    -0.66
    staking
    -0.63
     Job
    -0.60
     believer
    -0.60
    imon
    -0.58
     DV
    -0.57
     Cran
    -0.57
     hypocr
    -0.57
     depth
    -0.57
    isine
    -0.57
    POSITIVE LOGITS
    gie
    0.77
    ezvous
    0.71
    isters
    0.69
    hee
    0.68
     Mechdragon
    0.68
    ª
    0.68
    entimes
    0.67
    ggle
    0.67
    rats
    0.67
    æ©Ł
    0.65
    Activations Density 0.000%

    No Known Activations