© 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 24576
    9. 8-RES_FS24576-JB
    11. 12702
    Prev
    Next
    INDEX
    Explanations

    technical terms related to gaming and rule sets

    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
    393
    +0.16
    0.8%
    362
    +0.12
    0.6%
    167
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.16
    0.12
    393
    +0.12
    0.09
    385
    +0.12
    0.08
    Negative Logits
    vernment
    -0.87
    SPONSORED
    -0.73
     anew
    -0.71
    ghazi
    -0.65
    failed
    -0.64
     Carmen
    -0.62
    Britain
    -0.61
     Latvia
    -0.60
    LOCK
    -0.60
    andise
    -0.59
    POSITIVE LOGITS
     slightest
    1.28
     latter
    1.23
     simplest
    1.16
     widest
    1.04
     aforementioned
    0.96
     same
    0.96
    ses
    0.95
     majority
    0.94
     smallest
    0.93
     foregoing
    0.93
    Activations Density 0.446%

    No Known Activations