Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 10-RES-JB
    6. 17198
    Prev
    Next
    INDEX
    Explanations

    mentions of medical professions, specifically doctors and physicians, as well as actions related to medical care such as prescribing and examining patients

    oai_token-act-pair · gpt-3.5-turbo

    references to medical professionals, particularly doctors and physicians

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Territories
    -0.90
     Flavoring
    -0.66
     Strongh
    -0.66
    ndra
    -0.66
    rw
    -0.63
    wm
    -0.61
     Dwell
    -0.61
     lyr
    -0.61
     Dwar
    -0.61
    footed
    -0.59
    POSITIVE LOGITS
     prescribing
    1.25
     prescribe
    1.04
     specializing
    1.03
     treating
    1.00
     doctor
    0.99
     diagn
    0.96
     presc
    0.96
    doctor
    0.96
     clinic
    0.96
     diagnoses
    0.96
    Activations Density 0.071%

    No Known Activations