© 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 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. 10-RES-JB
    11. 6156
    Prev
    Next
    INDEX
    Explanations

    words related to spiritual or ethereal concepts

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of "spirit" in various contexts

    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
    theless
    -0.79
     Abstract
    -0.66
    upon
    -0.66
    enegger
    -0.63
    MS
    -0.63
     CVE
    -0.62
     WATCHED
    -0.61
     Kessler
    -0.61
     Synd
    -0.61
     Printed
    -0.60
    POSITIVE LOGITS
    uality
    1.49
    ual
    1.09
    uous
    1.03
    ually
    0.97
    uel
    0.82
    hearted
    0.81
    UAL
    0.79
    isphere
    0.75
    atics
    0.74
    stones
    0.74
    Activations Density 0.017%

    No Known Activations