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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 0-RES-JB
    6. 17339
    Prev
    Next
    INDEX
    Explanations

    instances of the word "restroom"

    oai_token-act-pair · gpt-3.5-turbo

    references to restrooms and associated policies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     nails
    -0.89
     armour
    -0.85
     Infinity
    -0.85
    colour
    -0.80
     fibre
    -0.77
     colours
    -0.76
     realised
    -0.75
     maths
    -0.73
     aluminium
    -0.71
     UK
    -0.69
    POSITIVE LOGITS
     restrooms
    2.90
     restroom
    2.57
     viewpoint
    1.30
     bathrooms
    1.30
    orer
    1.10
     seating
    1.07
     odor
    1.06
    ENN
    1.01
    rouse
    0.97
     prostitution
    0.93
    Activations Density 0.042%

    No Known Activations