© 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. 9-RES-JB
    11. 3089
    Prev
    Next
    INDEX
    Explanations

    male pronouns associated with medical situations being rushed to hospitals or suffering injuries

    oai_token-act-pair · gpt-3.5-turbo

    references to individuals and their actions in a narrative context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •male

    No Comments

    Negative Logits
     awarding
    -0.76
    ablishment
    -0.72
     prescribing
    -0.69
    dp
    -0.68
     recommending
    -0.68
    é¾įå¥ij士
    -0.67
     Publishing
    -0.66
     endorsements
    -0.65
     Creating
    -0.65
     appointing
    -0.63
    POSITIVE LOGITS
     underwent
    1.40
     suffered
    1.30
     survived
    1.29
     disappeared
    1.27
     escaped
    1.24
     died
    1.22
     wandered
    1.17
     vanished
    1.13
     fled
    1.10
     suffers
    1.09
    Activations Density 0.521%

    No Known Activations