© 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. 12366
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to political figures and locations

    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
    712
    +0.11
    0.5%
    659
    +0.11
    0.5%
    428
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.11
    0.07
    284
    +0.11
    0.05
    232
    +0.10
    0.05
    Negative Logits
    itiveness
    -0.62
    rontal
    -0.61
     Dealer
    -0.59
    rocal
    -0.59
    prem
    -0.58
     Freak
    -0.58
    akura
    -0.56
     elig
    -0.56
     hereafter
    -0.55
    Filename
    -0.55
    POSITIVE LOGITS
    steen
    0.94
    phia
    0.87
    ande
    0.75
    ith
    0.75
    ando
    0.70
    endi
    0.70
    jah
    0.67
    ieu
    0.67
    stad
    0.67
    umen
    0.66
    Activations Density 0.260%

    No Known Activations