© 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. 14929
    Prev
    Next
    INDEX
    Explanations

    words related to clothing and fit, particularly focusing on the fit and comfort of bras

    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
    678
    +0.16
    0.7%
    699
    +0.12
    0.5%
    500
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    678
    +0.16
    0.05
    435
    +0.12
    0.04
    260
    +0.11
    0.04
    Negative Logits
    issance
    -0.84
    yssey
    -0.82
    istani
    -0.81
    capital
    -0.78
    agame
    -0.78
    ensable
    -0.77
    veyard
    -0.76
    emonic
    -0.75
    ongo
    -0.75
    atoon
    -0.73
    POSITIVE LOGITS
     diameter
    1.23
     circumference
    1.08
     casing
    1.04
    most
    0.98
     torso
    0.97
     rim
    0.96
     cavity
    0.96
     halves
    0.93
     hull
    0.93
     width
    0.92
    Activations Density 0.102%

    No Known Activations