© 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. 24302
    Prev
    Next
    INDEX
    Explanations

    the word "normal" as a predominant feature

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "normal" and its variations in different contexts

    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
    hani
    -0.72
    artisan
    -0.69
    better
    -0.66
    iosyncr
    -0.64
    Lens
    -0.63
    Hop
    -0.63
     Goff
    -0.62
    raped
    -0.62
    lest
    -0.61
    Sov
    -0.61
    POSITIVE LOGITS
    cy
    1.46
    ization
    1.32
    izes
    1.31
    izing
    1.31
    ised
    1.29
    isation
    1.24
    ize
    1.20
    ized
    1.16
    izer
    1.11
    izers
    1.11
    Activations Density 0.038%

    No Known Activations