© 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. 0-RES-JB
    11. 15174
    Prev
    Next
    INDEX
    Explanations

    mentions of the platform "Reddit" in the text

    oai_token-act-pair · gpt-3.5-turbo

    mentions of various social media platforms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     lost
    -0.75
     Wolf
    -0.73
     Ware
    -0.72
     Lost
    -0.71
     loss
    -0.71
     Loss
    -0.70
     Targ
    -0.70
     Nil
    -0.69
     Wind
    -0.69
     End
    -0.69
    POSITIVE LOGITS
    Reddit
    3.29
    Pinterest
    2.97
    LinkedIn
    2.26
    Tumblr
    2.24
    Print
    1.69
    Pocket
    1.63
    reddit
    1.48
    Twitter
    1.42
    Redditor
    1.37
    Facebook
    1.28
    Activations Density 0.027%

    No Known Activations