© 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. 9-RES-JB
    11. 7133
    Prev
    Next
    INDEX
    Explanations

    words related to documents, like "copy" and "excerpt."

    oai_token-act-pair · gpt-3.5-turbo

    references to items or excerpts from documents and materials

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rio
    -0.83
    poke
    -0.83
    nels
    -0.78
    hattan
    -0.77
    otics
    -0.76
    ghazi
    -0.74
    cosystem
    -0.73
    gpu
    -0.73
    »Ĵ
    -0.73
    hesda
    -0.73
    POSITIVE LOGITS
     Quran
    1.09
     Koran
    1.04
     scripture
    0.98
     excerpts
    0.97
     manuscript
    0.96
     poems
    0.95
     Psal
    0.92
     manuscripts
    0.92
     Qur
    0.89
     poetry
    0.89
    Activations Density 0.176%

    No Known Activations