© 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. 20293
    Prev
    Next
    INDEX
    Explanations

    phrases or words indicating a significant achievement or historical milestone

    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
    373
    +0.14
    0.7%
    185
    +0.11
    0.5%
    691
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    21
    +0.14
    0.04
    366
    +0.11
    0.04
    5
    +0.11
    0.03
    Negative Logits
    maps
    -0.78
    Õ
    -0.73
    lag
    -0.70
    à©
    -0.70
    redits
    -0.69
    licks
    -0.67
    deals
    -0.64
     Rooms
    -0.63
    acies
    -0.63
    asions
    -0.62
    POSITIVE LOGITS
     beneficiary
    1.14
     casualty
    1.09
     person
    1.07
     recipient
    1.06
     beneficiaries
    0.96
     foreigner
    0.91
     baseman
    0.88
     responders
    0.88
     participant
    0.88
     contestant
    0.86
    Activations Density 0.083%

    No Known Activations