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. 0-RES-JB
    6. 10516
    Prev
    Next
    INDEX
    Explanations

    keywords related to computer desktop applications

    oai_token-act-pair · gpt-3.5-turbo

    references to desktop environments or applications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ogh
    -0.75
     Guerrero
    -0.74
     Boyle
    -0.72
     Rash
    -0.71
     rab
    -0.70
     Oliv
    -0.70
    Neg
    -0.69
    Fer
    -0.69
     Irving
    -0.67
     Vacc
    -0.66
    POSITIVE LOGITS
     desktop
    3.40
     Desktop
    2.68
    desktop
    2.42
    Desktop
    2.01
     desk
    1.70
     laptop
    1.51
     tablet
    1.43
     tabletop
    1.38
     laptops
    1.33
     KDE
    1.28
    Activations Density 0.018%

    No Known Activations