© 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. 0-RES-JB
    11. 12690
    Prev
    Next
    INDEX
    Explanations

    repeated actions or instructions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "repeat" and related concepts in instructional or procedural content

    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
     esp
    -0.77
     impl
    -0.73
     ed
    -0.65
     tele
    -0.63
     bull
    -0.63
     indebted
    -0.63
     fences
    -0.62
     cens
    -0.62
     Dillon
    -0.62
     sil
    -0.62
    POSITIVE LOGITS
     Repeat
    2.88
    Repeat
    2.23
    repeat
    1.29
    alogy
    1.22
    omon
    1.11
    alog
    1.01
    Repe
    0.99
    romy
    0.97
     rehearsal
    0.95
     reuse
    0.94
    Activations Density 0.059%

    No Known Activations