© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 4636
    Prev
    Next
    INDEX
    Explanations

    situations involving rules and consequences, possibly related to law or guidelines

    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
    551
    +0.16
    0.8%
    60
    +0.10
    0.5%
    757
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    551
    +0.16
    0.09
    60
    +0.10
    0.05
    757
    +0.09
    0.05
    Negative Logits
     Seeking
    -0.66
     bidding
    -0.64
     Fighting
    -0.63
     Mant
    -0.61
     Fighter
    -0.61
     submitting
    -0.60
     honoring
    -0.60
    bats
    -0.58
     Generation
    -0.57
     hating
    -0.57
    POSITIVE LOGITS
    't
    1.56
    adian
    1.22
    berra
    1.15
    isters
    1.14
    NOT
    1.08
     easily
    1.06
    ister
    1.05
     be
    1.01
     afford
    0.95
     withstand
    0.93
    Activations Density 0.138%

    No Known Activations