© 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. 10-RES-JB
    11. 5387
    Prev
    Next
    INDEX
    Explanations

    numerical measurements and specific technical terms

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks and their various placements within sentences

    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
     downright
    -0.94
     popping
    -0.92
     tack
    -0.87
     cringe
    -0.81
     shaming
    -0.81
     tempt
    -0.80
     insanely
    -0.79
     creeping
    -0.78
     painfully
    -0.78
     surely
    -0.78
    POSITIVE LOGITS
     During
    1.51
     Approximately
    1.49
     Afterwards
    1.44
     Additionally
    1.43
     Each
    1.39
     Following
    1.36
     Throughout
    1.34
     However
    1.33
     Upon
    1.33
     Initially
    1.32
    Activations Density 0.378%

    No Known Activations