© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 10135
    Prev
    Next
    INDEX
    Explanations

    medical and biological terms related to bodily fluids such as urine, saliva, and blood

    oai_token-act-pair · gpt-3.5-turbo

    references to urine and related biological samples

    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
    aido
    -0.81
    ovie
    -0.80
    Offline
    -0.75
    arent
    -0.74
     Cosponsors
    -0.74
    opher
    -0.73
    Joined
    -0.72
     impunity
    -0.70
    neg
    -0.70
    awaru
    -0.69
    POSITIVE LOGITS
     urine
    0.89
    pee
    0.81
     samples
    0.79
    ysis
    0.78
     gland
    0.76
     leakage
    0.76
     microbiome
    0.75
     ejac
    0.74
     bladder
    0.73
     emission
    0.73
    Activations Density 0.011%

    No Known Activations