© 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 EvalsExportsAPI 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. 17873
    Prev
    Next
    INDEX
    Explanations

    preferences and tendencies towards particular humor styles and political alignments

    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
    726
    +0.12
    0.5%
    276
    +0.11
    0.5%
    218
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    276
    +0.12
    0.01
    726
    +0.11
    0.01
    222
    +0.11
    0.01
    Negative Logits
    mberg
    -0.80
    manship
    -0.68
    mans
    -0.67
    Sus
    -0.67
    wordpress
    -0.66
    tics
    -0.63
    VERSION
    -0.63
    meal
    -0.62
    WARN
    -0.62
    ctica
    -0.62
    POSITIVE LOGITS
     preferences
    1.02
     favoring
    0.90
     preference
    0.82
    eering
    0.81
     favoured
    0.80
    ministic
    0.78
     pane
    0.77
     favored
    0.76
    ļéĨĴ
    0.76
     choices
    0.74
    Activations Density 0.015%

    No Known Activations