© 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. 3-RES-JB
    11. 20629
    Prev
    Next
    INDEX
    Explanations

    references to the platform 'Reddit'

    oai_token-act-pair · gpt-3.5-turbo

    the mention of the platform Reddit

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ³³³³³³³³³³³³³³³³
    -0.74
    ³³³³³³³³
    -0.72
     CHO
    -0.66
     accur
    -0.63
     MPG
    -0.63
     Kissinger
    -0.61
     Thom
    -0.61
    tin
    -0.61
    bery
    -0.60
     FAT
    -0.60
    POSITIVE LOGITS
    Reddit
    1.17
    icum
    1.08
     Reddit
    1.00
    reddits
    1.00
    ors
    0.95
     Username
    0.95
    urous
    0.83
    Tumblr
    0.80
     thread
    0.79
     username
    0.77
    Activations Density 0.008%

    No Known Activations