© 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 EvalsExportsinterp-engineNEWAPI 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. 11-RES-JB
    11. 21163
    Prev
    Next
    INDEX
    Explanations

    references to the use of paper in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    elements related to artistic representation and visual media

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    FML
    -0.88
     Marketplace
    -0.87
    Season
    -0.80
    spawn
    -0.78
     Chern
    -0.77
     Cosponsors
    -0.74
     Raider
    -0.73
     Motorsport
    -0.73
    ¥µ
    -0.72
    \\\\
    -0.71
    POSITIVE LOGITS
     handwriting
    2.00
     typew
    1.87
     parchment
    1.82
     scrib
    1.80
     ink
    1.79
     handwritten
    1.77
     paper
    1.76
     pencil
    1.71
     photoc
    1.60
     notebook
    1.59
    Activations Density 0.571%

    No Known Activations