© 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. 11543
    Prev
    Next
    INDEX
    Explanations

    links and calls to action related to sharing or reading stories

    oai_token-act-pair · gpt-3.5-turbo

    links and call-to-action phrases related to sharing stories or signing up for newsletters

    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
     Classification
    -0.38
     dracon
    -0.38
    naires
    -0.35
    naire
    -0.34
    ,[
    -0.34
    alties
    -0.33
     positives
    -0.33
    RESULTS
    -0.33
     domest
    -0.32
    wards
    -0.31
    POSITIVE LOGITS
    iframe
    0.57
     embed
    0.45
    embed
    0.44
     UNCLASSIFIED
    0.42
     Email
    0.39
    News
    0.38
    aily
    0.38
    Subscribe
    0.37
    news
    0.37
     POLITICO
    0.37
    Activations Density 0.597%

    No Known Activations