© 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. 1-RES-JB
    11. 11483
    Prev
    Next
    INDEX
    Explanations

    references to gentlemen or gentleman-like behavior

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of gentleman

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Fuel
    -0.76
    onz
    -0.68
    aeda
    -0.67
    aya
    -0.65
    rises
    -0.65
    WT
    -0.64
    PT
    -0.64
     Printed
    -0.63
    inated
    -0.62
    helps
    -0.62
    POSITIVE LOGITS
     gentleman
    1.46
     gentlemen
    1.26
     Gentleman
    0.98
    entle
    0.87
    friend
    0.75
     Compan
    0.75
    cade
    0.71
     introdu
    0.71
    gery
    0.70
     Gentle
    0.70
    Activations Density 0.005%

    No Known Activations