© 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. 11624
    Prev
    Next
    INDEX
    Explanations

    mentions of social media posts or updates

    oai_token-act-pair · gpt-3.5-turbo

    instances of social media interactions or references

    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
     remains
    -0.72
     arrives
    -0.71
    JUST
    -0.67
     accordingly
    -0.66
    oeuv
    -0.64
     ensures
    -0.64
     thereafter
    -0.64
     stays
    -0.63
     acknowledges
    -0.63
     insists
    -0.62
    POSITIVE LOGITS
     reddit
    0.80
     Patreon
    0.80
     Discord
    0.75
     subreddit
    0.75
     facebook
    0.73
     Lovecraft
    0.72
     github
    0.72
     Github
    0.72
     AMA
    0.70
     myster
    0.70
    Activations Density 0.982%

    No Known Activations