© 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. 18594
    Prev
    Next
    INDEX
    Explanations

    social media-related terms such as likes, shares, views, and followers

    oai_token-act-pair · gpt-3.5-turbo

    mentions of social media engagement and metrics

    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
     laboratories
    -0.79
     deregulation
    -0.79
     conductor
    -0.77
     Quart
    -0.76
     Shelter
    -0.74
     Auditor
    -0.72
     McCoy
    -0.71
     Siem
    -0.70
     contingency
    -0.69
    ederation
    -0.68
    POSITIVE LOGITS
     retweet
    1.62
    votes
    1.47
     replies
    1.43
     Likes
    1.39
     posts
    1.37
    reply
    1.31
     clicks
    1.25
     subreddits
    1.23
     spam
    1.22
    comments
    1.20
    Activations Density 0.638%

    No Known Activations