© 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 49152
    9. 8-RES_FS49152-JB
    11. 25971
    Prev
    Next
    INDEX
    Explanations

    references to specific names or titles

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    339
    +0.12
    0.5%
    320
    +0.11
    0.5%
    148
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    48
    +0.12
    0.01
    232
    +0.11
    0.01
    343
    +0.10
    0.01
    Negative Logits
    ORGE
    -0.75
     steroids
    -0.60
     TYPE
    -0.59
    PsyNetMessage
    -0.58
    listed
    -0.57
     removable
    -0.57
     desk
    -0.57
     steroid
    -0.56
     AAP
    -0.56
    DonaldTrump
    -0.56
    POSITIVE LOGITS
    mire
    1.09
    achev
    1.04
    atche
    0.95
    asus
    0.95
    atell
    0.90
    achu
    0.89
    sburgh
    0.87
    lia
    0.86
    rint
    0.84
    atoon
    0.83
    Activations Density 0.007%

    No Known Activations