© 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. 8-RES-JB
    11. 2202
    Prev
    Next
    INDEX
    Explanations

    specific titles and ranks within religious and law enforcement organizations

    oai_token-act-pair · gpt-3.5-turbo

    titles or positions of authority, particularly related to religious and law enforcement figures

    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
     tumble
    -0.84
     printing
    -0.73
     fracture
    -0.72
     panc
    -0.65
     simmer
    -0.65
     ballet
    -0.64
     comparable
    -0.64
     drive
    -0.64
     captcha
    -0.64
     backward
    -0.63
    POSITIVE LOGITS
    essors
    0.99
     Timothy
    0.93
    herty
    0.90
    .,
    0.90
     Joseph
    0.88
     Maj
    0.88
     Lt
    0.87
     Sgt
    0.86
    adier
    0.85
    Lt
    0.85
    Activations Density 0.069%

    No Known Activations