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

    Neuronpedia

    Jacobian LensNEW
    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. 1991
    Prev
    Next
    INDEX
    Explanations

    words or acronyms with "Ds" or "Bs" or "Gs" or "Es" or "Fs" as well as specific features or actions related to them

    oai_token-act-pair · gpt-3.5-turbo

    abbreviations or acronyms related to specific categories or classifications, particularly in a technical or creative context

    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
    emark
    -0.82
     SOU
    -0.64
    weight
    -0.63
    court
    -0.62
    ĸļ
    -0.62
     PR
    -0.60
    union
    -0.60
    law
    -0.60
    mus
    -0.59
    pun
    -0.59
    POSITIVE LOGITS
    Ds
    2.88
    Cs
    2.82
    Bs
    2.46
    Gs
    2.46
    Fs
    2.20
    Es
    2.08
    Rs
    2.00
    Ns
    1.91
    Ts
    1.83
    Ws
    1.79
    Activations Density 0.035%

    No Known Activations