© 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. 8-RES-JB
    11. 22535
    Prev
    Next
    INDEX
    Explanations

    references to legal proceedings and court-related terms

    oai_token-act-pair · gpt-3.5-turbo

    references to various courts and legal proceedings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ku
    -0.72
    ec
    -0.69
    eco
    -0.67
    ulic
    -0.66
    zsche
    -0.66
    edia
    -0.66
    aned
    -0.66
    yip
    -0.65
    eele
    -0.64
    jiang
    -0.63
    POSITIVE LOGITS
     judge
    1.05
     Judge
    1.03
    rooms
    0.98
     magistrate
    0.97
     Clerk
    0.90
     Settlement
    0.86
     proceedings
    0.86
    auld
    0.85
    istrate
    0.83
    Court
    0.82
    Activations Density 0.064%

    No Known Activations