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. 3-RES-JB
    6. 7543
    Prev
    Next
    INDEX
    Explanations

    legal terms and concepts

    oai_token-act-pair · gpt-3.5-turbo

    references to legal terms and constitutional clauses

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     highs
    -0.81
     Seah
    -0.79
     Bild
    -0.78
    icro
    -0.78
     realise
    -0.77
     Hai
    -0.77
    ellery
    -0.75
    ufact
    -0.73
    anie
    -0.72
    ongh
    -0.72
    POSITIVE LOGITS
     justices
    1.17
    Attorney
    1.15
     unconstitutional
    1.13
     appellate
    1.08
     attorneys
    1.05
    Justice
    1.04
     Scalia
    1.03
     plaintiffs
    1.02
     constitutional
    1.01
     statute
    1.00
    Activations Density 0.737%

    No Known Activations