© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 15336
    Prev
    Next
    INDEX
    Explanations

    mentions or references to specific Twitter usernames

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    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
    Neuron Alignment
    Index
    Value
    % of L₁
    160
    +0.15
    0.7%
    311
    +0.12
    0.6%
    107
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    107
    +0.15
    0.02
    160
    +0.12
    0.02
    691
    +0.12
    0.02
    Negative Logits
     ply
    -0.73
    pload
    -0.69
     bounce
    -0.68
     Turing
    -0.68
     screenings
    -0.67
     chlorine
    -0.67
     repent
    -0.66
     divers
    -0.66
     ric
    -0.65
     hardened
    -0.65
    POSITIVE LOGITS
    ebook
    1.28
    dust
    1.08
    official
    1.00
    tro
    0.97
    chance
    0.97
    user
    0.95
    pill
    0.94
    medium
    0.94
    blank
    0.94
    must
    0.93
    Activations Density 0.017%

    No Known Activations