© 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. 4368
    Prev
    Next
    INDEX
    Explanations

    words related to illnesses, specifically different types of flu

    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
    0
    +0.13
    0.6%
    78
    +0.09
    0.4%
    211
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    0
    +0.13
    0.01
    78
    +0.09
    0.01
    211
    +0.09
    0.01
    Negative Logits
     Skydragon
    -0.90
     Fathers
    -0.81
     Limits
    -0.80
     guiActiveUnfocused
    -0.73
     Pwr
    -0.68
     Blessed
    -0.67
     Phar
    -0.67
     Parties
    -0.66
     Universe
    -0.66
     Cosmos
    -0.65
    POSITIVE LOGITS
    orescence
    1.47
    oresc
    1.43
    ency
    1.21
    enza
    1.18
    ently
    1.09
    orescent
    1.08
    encies
    1.06
    iencies
    1.06
    ipeg
    1.04
    uates
    1.03
    Activations Density 0.019%

    No Known Activations