© 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. 17122
    Prev
    Next
    INDEX
    Explanations

    symbols or icons representing nations or cultures

    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
    447
    +0.16
    0.7%
    237
    +0.13
    0.6%
    724
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.16
    0.04
    237
    +0.13
    0.03
    608
    +0.12
    0.04
    Negative Logits
    neau
    -0.82
    isode
    -0.81
    hari
    -0.80
    zon
    -0.75
    asus
    -0.73
     briefed
    -0.73
    asive
    -0.71
    eworks
    -0.69
    BILL
    -0.68
    matched
    -0.67
    POSITIVE LOGITS
     masculinity
    1.18
     purity
    1.09
     bravery
    1.01
     eternity
    1.01
     patriotism
    0.98
     individuality
    0.98
     virginity
    0.95
     innocence
    0.93
     reverence
    0.92
     modesty
    0.91
    Activations Density 0.148%

    No Known Activations