© 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. 8-RES-JB
    11. 17734
    Prev
    Next
    INDEX
    Explanations

    names of politicians or public figures

    oai_token-act-pair · gpt-3.5-turbo

    names of public figures and events in a political context

    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
    Contents
    -0.76
    strument
    -0.73
    material
    -0.70
    ÙĴ
    -0.69
     ().
    -0.69
    $.
    -0.68
    common
    -0.68
    Ire
    -0.66
     Morty
    -0.65
    Donnell
    -0.65
    POSITIVE LOGITS
     reacts
    1.17
     slams
    1.15
     Says
    1.09
     defends
    1.07
     shuts
    1.06
     approves
    1.02
     warns
    1.01
     refuses
    0.99
     denies
    0.99
     responds
    0.97
    Activations Density 0.502%

    No Known Activations