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. 11-RES-JB
    6. 19353
    Prev
    Next
    INDEX
    Explanations

    instances of the word "married" followed by the name of a person

    oai_token-act-pair · gpt-3.5-turbo

    verbs related to actions involving relationships and legal or formal processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Interstitial
    -0.85
    catentry
    -0.82
    wcs
    -0.76
    padding
    -0.71
    Supplement
    -0.71
    alde
    -0.70
    fal
    -0.70
    wcsstore
    -0.70
    ItemThumbnailImage
    -0.69
    amins
    -0.68
    POSITIVE LOGITS
     fellow
    0.94
     former
    0.93
     Jonathan
    0.92
     Lt
    0.91
     David
    0.89
     Denis
    0.89
     incumbent
    0.88
     longtime
    0.88
     Mehran
    0.85
     Lieutenant
    0.85
    Activations Density 0.300%

    No Known Activations