© 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. 8-RES-JB
    11. 12566
    Prev
    Next
    INDEX
    Explanations

    phrases starting with special characters followed by numbers

    oai_token-act-pair · gpt-3.5-turbo

    symbols or special characters, indicating potential emphasis or urgency in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     dispers
    -0.77
     shack
    -0.71
     orally
    -0.69
     range
    -0.67
     Belg
    -0.67
     assemb
    -0.66
     machine
    -0.65
     confinement
    -0.65
     contraception
    -0.65
     Slov
    -0.64
    POSITIVE LOGITS
    ª
    1.41
    ¤
    1.38
    ¹
    1.33
    į
    1.32
    Ń
    1.29
    ľ
    1.23
    ¡
    1.22
    Ķ
    1.21
    ´
    1.20
    ĺ
    1.20
    Activations Density 0.103%

    No Known Activations