© 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. 21211
    Prev
    Next
    INDEX
    Explanations

    pronouns indicating specific individuals, particularly 'she' and 'he'

    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
    762
    +0.12
    0.6%
    126
    +0.12
    0.5%
    681
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    354
    +0.12
    0.07
    762
    +0.12
    0.07
    566
    +0.11
    0.06
    Negative Logits
     Globe
    -0.83
     Concord
    -0.70
     Mile
    -0.66
     Akron
    -0.65
     Kirin
    -0.64
     Kinn
    -0.63
     Dusk
    -0.62
     Torrent
    -0.62
     Gerard
    -0.62
     Hunting
    -0.61
    POSITIVE LOGITS
    'll
    1.37
    'd
    1.26
    're
    1.21
    've
    1.16
     could
    1.06
    'm
    0.95
     wouldn
    0.93
    arers
    0.90
     should
    0.88
     faked
    0.87
    Activations Density 0.221%

    No Known Activations