© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 20451
    Prev
    Next
    INDEX
    Explanations

    technical descriptions of mechanical components and devices

    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
    480
    +0.48
    2.4%
    678
    +0.15
    0.7%
    266
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.48
    0.01
    678
    +0.15
    0.06
    266
    +0.12
    0.07
    Negative Logits
     Rebell
    -0.72
    Jews
    -0.70
    efe
    -0.69
    ologne
    -0.68
     Hallow
    -0.68
    politics
    -0.66
     Scholar
    -0.65
    é»Ĵ
    -0.65
    omore
    -0.65
    Politics
    -0.64
    POSITIVE LOGITS
     removable
    1.09
     adjustable
    1.06
     modular
    1.05
     apparatus
    1.04
     interface
    1.03
     subsystem
    1.01
     circuitry
    1.01
     chassis
    1.00
     interfaces
    0.95
     auxiliary
    0.93
    Activations Density 0.781%

    No Known Activations