© 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. 10-RES-JB
    11. 7790
    Prev
    Next
    INDEX
    Explanations

    proper nouns or names in a French context

    oai_token-act-pair · gpt-3.5-turbo

    references to significant events or changes in France

    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
     Kush
    -1.06
     Boise
    -1.00
     Sasuke
    -0.99
     Mariners
    -0.98
     Kodi
    -0.96
     Naruto
    -0.94
    Seattle
    -0.93
    rower
    -0.90
     Tacoma
    -0.88
    Jake
    -0.88
    POSITIVE LOGITS
     French
    2.05
     France
    2.04
    France
    2.01
     Hollande
    1.97
     Marse
    1.94
    French
    1.93
     François
    1.89
     Paris
    1.81
    Paris
    1.79
     Ãī
    1.79
    Activations Density 0.557%

    No Known Activations