© 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. 17002
    Prev
    Next
    INDEX
    Explanations

    information related to populations and demographics

    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
    181
    +0.12
    0.6%
    126
    +0.12
    0.5%
    256
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.12
    -0.01
    260
    +0.12
    0.04
    655
    +0.11
    0.03
    Negative Logits
     grounding
    -0.71
     remedies
    -0.70
     whiff
    -0.70
     primer
    -0.69
     shader
    -0.68
     treadmill
    -0.68
     breeze
    -0.68
     deals
    -0.68
     rulings
    -0.67
     brake
    -0.67
    POSITIVE LOGITS
    population
    1.16
    Population
    1.09
     numbering
    1.00
    number
    0.98
    count
    0.97
     dwind
    0.95
    NUM
    0.92
     population
    0.92
     outnumbered
    0.89
    MAX
    0.88
    Activations Density 0.466%

    No Known Activations