© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 24035
    Prev
    Next
    INDEX
    Explanations

    occurrences of names or proper nouns, particularly those related to significant events or figures

    oai_token-act-pair · gpt-4o-miniTriggered 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
    407
    +0.12
    0.5%
    732
    +0.12
    0.5%
    574
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    392
    +0.12
    0.04
    232
    +0.12
    0.03
    659
    +0.11
    0.03
    Negative Logits
     intermedi
    -0.83
    interstitial
    -0.80
    agascar
    -0.71
     364
    -0.71
     Alexis
    -0.69
     INF
    -0.68
     Malone
    -0.68
     Memor
    -0.67
     Instr
    -0.67
     Mig
    -0.66
    POSITIVE LOGITS
    w
    1.23
    W
    1.17
    WM
    1.17
    wi
    1.10
    WC
    1.07
    WA
    1.07
    wl
    1.05
    wick
    1.05
    wer
    1.04
    wa
    1.03
    Activations Density 0.134%

    No Known Activations