© 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. 4-RES-JB
    11. 17184
    Prev
    Next
    INDEX
    Explanations

    software-related terms such as "GNOME", "Shell", "KDE", "Qt", and "Fedora"

    oai_token-act-pair · gpt-3.5-turbo

    references to software and programming environments related to GNOME and KDE

    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
    ilage
    -0.81
    rahim
    -0.79
    Condition
    -0.76
    indu
    -0.73
    ousing
    -0.72
    Indian
    -0.71
    ihil
    -0.71
    buy
    -0.71
    ought
    -0.70
    orders
    -0.70
    POSITIVE LOGITS
     KDE
    0.86
     GNOME
    0.84
     splend
    0.76
     Plasma
    0.76
     Shell
    0.74
     decoration
    0.73
     Orchestra
    0.73
     desktop
    0.71
     Forums
    0.68
     wrapper
    0.67
    Activations Density 0.010%

    No Known Activations