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. 1-RES-JB
    6. 13680
    Prev
    Next
    INDEX
    Explanations

    administrative-related terms and phrases

    oai_token-act-pair · gpt-3.5-turbo

    references to administrative processes or issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    âķIJâķIJ
    -0.84
    λ
    -0.83
    words
    -0.77
     Emin
    -0.75
    DonaldTrump
    -0.74
     Kinnikuman
    -0.73
    False
    -0.72
    hner
    -0.71
    ym
    -0.71
    pter
    -0.71
    POSITIVE LOGITS
     administrative
    1.17
    rative
    0.92
     administ
    0.91
     disciplinary
    0.90
     administrator
    0.90
     enqu
    0.86
     wizard
    0.84
     managerial
    0.81
     cas
    0.81
     procedures
    0.79
    Activations Density 0.008%

    No Known Activations