© 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. 22320
    Prev
    Next
    INDEX
    Explanations

    references to family relationships, particularly focusing on the father figure

    oai_token-act-pair · gpt-4o-miniTriggered 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
    218
    +0.16
    0.7%
    309
    +0.10
    0.4%
    53
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    207
    +0.16
    0.03
    49
    +0.10
    0.04
    218
    +0.09
    0.03
    Negative Logits
    estyles
    -0.73
     guiName
    -0.71
     cones
    -0.69
    nergy
    -0.68
    erning
    -0.66
    leck
    -0.66
    matically
    -0.66
     targeting
    -0.65
    dimensional
    -0.65
    antics
    -0.64
    POSITIVE LOGITS
     died
    0.95
     divorced
    0.92
     deceased
    0.92
     dementia
    0.85
     married
    0.83
     disappro
    0.82
     perished
    0.82
     diabetic
    0.79
     Alzheimer
    0.78
     retired
    0.76
    Activations Density 0.420%

    No Known Activations