© 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. 10963
    Prev
    Next
    INDEX
    Explanations

    terms related to corruption or unethical behavior

    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
    113
    +0.12
    0.6%
    339
    +0.10
    0.5%
    588
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.03
    190
    +0.10
    0.02
    588
    +0.10
    0.02
    Negative Logits
    gap
    -0.85
     Anxiety
    -0.78
    ruck
    -0.73
    joice
    -0.72
    iphany
    -0.69
    zig
    -0.68
    fleet
    -0.68
    gain
    -0.67
    oleon
    -0.67
    plane
    -0.67
    POSITIVE LOGITS
     dealings
    0.87
    ly
    0.85
    ibly
    0.83
    ible
    0.80
    ulent
    0.79
    nesses
    0.78
    NESS
    0.78
    ingly
    0.74
    glers
    0.72
    ness
    0.72
    Activations Density 0.059%

    No Known Activations