© 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. 11-RES-JB
    11. 943
    Prev
    Next
    INDEX
    Explanations

    verbal expressions related to privilege, honor, and participation in projects or programs

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing participation or involvement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     dunno
    -0.84
     dislike
    -0.79
     gimmick
    -0.68
    xual
    -0.67
     lame
    -0.67
     bitch
    -0.66
     downgrade
    -0.66
     bluff
    -0.65
    avoid
    -0.64
     apocalypse
    -0.62
    POSITIVE LOGITS
     integral
    1.20
     instrumental
    1.20
     part
    1.17
     pivotal
    0.99
     invaluable
    0.96
     intimately
    0.95
     pioneers
    0.94
     indispensable
    0.92
     stewards
    0.92
     involved
    0.90
    Activations Density 0.334%

    No Known Activations