© 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. 3101
    Prev
    Next
    INDEX
    Explanations

    mentions of controversial events or issues in politics and society

    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
    157
    +0.13
    0.6%
    581
    +0.09
    0.4%
    534
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    157
    +0.13
    0.08
    8
    +0.09
    0.06
    55
    +0.09
    0.07
    Negative Logits
     fixme
    -0.70
    atos
    -0.69
    bowl
    -0.67
     ILCS
    -0.64
    ula
    -0.63
    oji
    -0.62
    Joined
    -0.61
    alde
    -0.61
     [|
    -0.60
    si
    -0.60
    POSITIVE LOGITS
     nonetheless
    1.23
    etheless
    1.08
     nevertheless
    0.91
     strangely
    0.74
     curiously
    0.72
     undeniably
    0.70
     doubts
    0.69
     decidedly
    0.68
     remarkably
    0.68
     caution
    0.68
    Activations Density 0.795%

    No Known Activations