© 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. 1908
    Prev
    Next
    INDEX
    Explanations

    names of individuals

    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
    420
    +0.11
    0.5%
    595
    +0.10
    0.5%
    74
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    452
    +0.11
    0.05
    46
    +0.10
    0.05
    461
    +0.09
    0.04
    Negative Logits
     NRL
    -0.68
     understatement
    -0.63
    ishable
    -0.61
    cientious
    -0.55
     Corpus
    -0.54
     poppy
    -0.54
    Tes
    -0.54
    ween
    -0.54
     Tenn
    -0.53
     satisfaction
    -0.53
    POSITIVE LOGITS
     specializes
    1.00
     consisted
    0.97
    enegger
    0.94
     testified
    0.86
     died
    0.85
     reportedly
    0.81
     was
    0.80
     wore
    0.79
     survived
    0.79
     lasted
    0.78
    Activations Density 0.316%

    No Known Activations