© 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. 3688
    Prev
    Next
    INDEX
    Explanations

    phrases related to getting rid of or managing costumes

    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
    742
    +0.11
    0.5%
    527
    +0.09
    0.4%
    678
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    742
    +0.11
    0.01
    232
    +0.09
    0.01
    503
    +0.09
    0.01
    Negative Logits
    REDACTED
    -0.83
    DEN
    -0.75
    enegger
    -0.72
    BACK
    -0.66
     PowerPoint
    -0.65
    lace
    -0.63
    KEN
    -0.62
    BALL
    -0.61
     Dragonbound
    -0.61
     poisoned
    -0.60
    POSITIVE LOGITS
    ongs
    1.23
    ashing
    1.21
    ifty
    1.21
    isted
    1.02
    ifts
    1.02
    ashes
    1.02
    ifted
    1.01
    ift
    1.01
    ifting
    1.00
    umps
    0.99
    Activations Density 0.018%

    No Known Activations