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. 10466
    Prev
    Next
    INDEX
    Explanations

    phrases containing the prefix "non-" followed by a hyphen and a word

    oai_token-act-pair · gpt-3.5-turbo

    references to non-white individuals or groups

    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
     fallacy
    -0.88
     downfall
    -0.87
     mock
    -0.82
     dilemma
    -0.81
     ridicule
    -0.81
     folly
    -0.81
     undo
    -0.80
     disguise
    -0.80
     dossier
    -0.78
     caveat
    -0.77
    POSITIVE LOGITS
    citizens
    1.55
    white
    1.45
    members
    1.38
    resident
    1.38
    humans
    1.36
    Americans
    1.35
    wh
    1.35
    immigrant
    1.33
    married
    1.31
    profits
    1.31
    Activations Density 0.067%

    No Known Activations