© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 0-RES-JB
    11. 18144
    Prev
    Next
    INDEX
    Explanations

    the word 'rep' followed by a word or syllable

    oai_token-act-pair · gpt-3.5-turbo

    variations of the word "reparation."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ahar
    -0.76
    eday
    -0.75
    enario
    -0.68
    agne
    -0.68
     TAMADRA
    -0.68
    lass
    -0.67
     Fighters
    -0.67
    sing
    -0.65
    STEM
    -0.64
    inders
    -0.64
    POSITIVE LOGITS
     rep
    3.75
    rep
    1.68
     Rep
    1.65
    Rep
    1.48
     reps
    1.38
     repro
    1.36
     repe
    1.29
     Reps
    1.16
     repr
    1.16
     recomp
    1.15
    Activations Density 0.022%

    No Known Activations