© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 11977
    Prev
    Next
    INDEX
    Explanations

    references to physical locations, specifically galleries

    oai_token-act-pair · gpt-3.5-turbo

    references to galleries

    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
    Boost
    -0.70
     Elim
    -0.69
    ord
    -0.65
     internally
    -0.62
     Effective
    -0.61
    Commun
    -0.60
    arettes
    -0.60
    ordinate
    -0.60
    illegal
    -0.59
    phas
    -0.59
    POSITIVE LOGITS
     gallery
    4.24
     galleries
    3.27
     Gallery
    2.65
    gallery
    2.33
     slideshow
    2.09
    allery
    1.98
    Gallery
    1.85
     museum
    1.30
     curator
    1.24
     exhibition
    1.21
    Activations Density 0.006%

    No Known Activations