© 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. 3293
    Prev
    Next
    INDEX
    Explanations

    reversals, announcements, disclosures, and unexpected news events in the context of institutions and scandal

    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
    447
    +0.15
    0.7%
    189
    +0.10
    0.4%
    679
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    341
    +0.15
    0.05
    189
    +0.10
    0.05
    679
    +0.10
    0.06
    Negative Logits
    Pros
    -0.77
     dying
    -0.65
     vomit
    -0.61
     sacrific
    -0.59
    ocrates
    -0.58
     invincible
    -0.58
     murdering
    -0.58
     Likes
    -0.58
     raping
    -0.57
    avis
    -0.57
    POSITIVE LOGITS
     coincides
    1.10
     underscores
    1.10
     coincided
    1.07
     prompted
    1.04
     sparked
    0.97
     comes
    0.97
     stemmed
    0.96
     represents
    0.92
     includes
    0.89
     came
    0.89
    Activations Density 0.278%

    No Known Activations