© 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. 15566
    Prev
    Next
    INDEX
    Explanations

    keywords related to current events, political conflicts, and governmental actions

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    87
    +0.26
    1.3%
    447
    +0.16
    0.8%
    481
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    224
    +0.26
    0.09
    146
    +0.16
    0.07
    461
    +0.11
    0.06
    Negative Logits
    $.
    -0.61
    ocamp
    -0.60
    tnc
    -0.55
     eternity
    -0.54
    âĶĢâĶĢâĶĢâĶĢ
    -0.54
    achment
    -0.54
    ãĥ¬
    -0.53
    Eva
    -0.53
    Occ
    -0.53
    7601
    -0.53
    POSITIVE LOGITS
     meanwhile
    1.42
     reacted
    1.20
     responded
    1.02
     applauded
    1.00
     countered
    0.96
     welcomed
    0.95
     congratulated
    0.94
     intervened
    0.94
     also
    0.94
     echoed
    0.92
    Activations Density 0.608%

    No Known Activations