© 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. 6328
    Prev
    Next
    INDEX
    Explanations

    phrases related to timing and conditions

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions and their usage, especially in phrases indicating conditions or relationships

    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
     Raider
    -0.62
     Theft
    -0.61
    PORT
    -0.61
     mockery
    -0.60
     Reno
    -0.59
    verse
    -0.59
     Sutherland
    -0.58
    Tes
    -0.58
     Reeves
    -0.58
    Ö¼
    -0.57
    POSITIVE LOGITS
    why
    1.09
     why
    1.03
     whence
    1.00
    rogens
    0.95
     WHY
    0.95
    how
    0.91
     how
    0.90
    rogen
    0.90
     WHERE
    0.89
    when
    0.89
    Activations Density 0.038%

    No Known Activations