Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 4-RES-JB
    6. 18935
    Prev
    Next
    INDEX
    Explanations

    references to human rights and international organizations

    oai_token-act-pair · gpt-3.5-turbo

    references to complex social issues, particularly concerning identity and understanding within context

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

    No Comments

    Negative Logits
     booze
    -0.78
     torment
    -0.72
     cursing
    -0.69
     ruining
    -0.69
     drunk
    -0.68
     goof
    -0.67
     raping
    -0.66
     drunken
    -0.65
     aven
    -0.65
     loaf
    -0.65
    POSITIVE LOGITS
     Fiscal
    0.73
     Cosponsors
    0.71
     DOE
    0.69
    ortium
    0.69
     Implementation
    0.67
     jointly
    0.67
    negie
    0.67
     Retrieved
    0.66
     Funds
    0.66
    trak
    0.65
    Activations Density 2.116%

    No Known Activations