© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 26578
    Prev
    Next
    INDEX
    Explanations

    mentions of creative or artistic work

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    420
    +0.12
    0.5%
    698
    +0.10
    0.4%
    175
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    698
    +0.12
    0.03
    420
    +0.10
    0.02
    140
    +0.10
    0.02
    Negative Logits
     Ukrain
    -0.98
    wcs
    -0.75
    kefeller
    -0.71
    Args
    -0.64
    ylon
    -0.63
     champagne
    -0.61
    idate
    -0.61
    angular
    -0.61
    rition
    -0.60
     Gord
    -0.60
    POSITIVE LOGITS
     ethic
    1.42
    flows
    1.37
    station
    1.24
    aday
    1.15
    manship
    1.10
    bench
    1.08
    horse
    0.98
    mates
    0.90
    forces
    0.89
    papers
    0.89
    Activations Density 0.031%

    No Known Activations