© 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 EvalsExportsinterp-engineNEWAPI 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. 8-RES-JB
    11. 12314
    Prev
    Next
    INDEX
    Explanations

    phrases related to opinions or statements made by individuals

    oai_token-act-pair · gpt-3.5-turbo

    statements and claims from public figures in political contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Course
    -0.72
     Located
    -0.71
     ILCS
    -0.70
     Derby
    -0.67
     Photograph
    -0.65
     Rolls
    -0.64
     Created
    -0.63
    aughtered
    -0.63
    Rated
    -0.61
     Keeper
    -0.61
    POSITIVE LOGITS
     clar
    0.95
     sarcast
    0.93
    rompt
    0.93
     acknow
    0.90
     apologizing
    0.89
     emphatically
    0.89
     bluntly
    0.89
     scathing
    0.88
     succinct
    0.88
     eloqu
    0.87
    Activations Density 0.581%

    No Known Activations