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. 22802
    Prev
    Next
    INDEX
    Explanations

    instances of the word "sue"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "sue" and its variations, particularly in legal contexts

    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
    generated
    -0.64
    çľ
    -0.62
    display
    -0.60
     Crowley
    -0.58
    EC
    -0.58
     Cad
    -0.57
     activity
    -0.57
     Companion
    -0.57
    EG
    -0.56
     Belt
    -0.56
    POSITIVE LOGITS
     sue
    3.81
     suing
    2.68
     sued
    2.44
     lawsuits
    1.59
     lawsuit
    1.49
     prosecute
    1.39
     litigation
    1.30
     plaintiff
    1.28
     lawy
    1.25
     plaintiffs
    1.22
    Activations Density 0.016%

    No Known Activations