© 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. 15507
    Prev
    Next
    INDEX
    Explanations

    information related to sports events and game results

    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
    281
    +0.14
    0.6%
    31
    +0.11
    0.5%
    48
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    281
    +0.14
    0.05
    533
    +0.11
    0.05
    668
    +0.10
    0.04
    Negative Logits
     memos
    -0.85
     Emails
    -0.83
    soType
    -0.81
    Leaks
    -0.80
    abytes
    -0.78
     dialog
    -0.77
    agency
    -0.76
    Wiki
    -0.76
     empathy
    -0.76
    forums
    -0.75
    POSITIVE LOGITS
     unbeaten
    1.32
     undefeated
    1.29
     defeated
    1.14
     victorious
    1.14
     relegation
    1.10
     winning
    1.06
     win
    1.04
     wins
    1.02
     rematch
    1.02
     rivals
    1.02
    Activations Density 0.592%

    No Known Activations