© 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. 10-RES-JB
    11. 18948
    Prev
    Next
    INDEX
    Explanations

    words related to substances like sugar (represented by 'su'), regardless of the context

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the prefix "su"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Ferdinand
    -0.60
    forth
    -0.58
     union
    -0.57
     Union
    -0.57
     owl
    -0.57
     Noir
    -0.57
     Weir
    -0.56
     Hitchcock
    -0.56
     Lyme
    -0.55
     Ahead
    -0.55
    POSITIVE LOGITS
    fficiency
    1.31
    nder
    1.18
    zanne
    1.18
    ction
    1.08
    prisingly
    1.06
    ffic
    1.04
    cess
    1.04
    ccess
    1.01
    prise
    1.01
    itor
    0.97
    Activations Density 0.012%

    No Known Activations