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. 6982
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal proceedings and court activities

    oai_token-act-pair · gpt-3.5-turbo

    references to legal proceedings or the court system

    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
    gans
    -0.70
    Availability
    -0.65
    ombies
    -0.64
    Loading
    -0.63
    ives
    -0.62
    anguages
    -0.62
    rences
    -0.62
     Ahead
    -0.61
    -----
    -0.61
     (#
    -0.61
    POSITIVE LOGITS
     court
    3.75
     Court
    2.74
    Court
    2.67
    court
    2.67
     courts
    2.61
     COURT
    2.46
     Courts
    2.22
     courtroom
    2.13
     tribunal
    1.85
     judge
    1.78
    Activations Density 0.020%

    No Known Activations