© 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 EvalsExportsinterp-engineNEWAPI 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. 5162
    Prev
    Next
    INDEX
    Explanations

    action verbs associated with physical or intense activities

    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
    184
    +0.12
    0.6%
    373
    +0.12
    0.6%
    674
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.12
    0.05
    750
    +0.12
    0.05
    665
    +0.10
    0.06
    Negative Logits
     namely
    -0.58
     requiring
    -0.55
     characteristic
    -0.55
    Solution
    -0.55
     deterrent
    -0.55
     evident
    -0.55
     Due
    -0.55
     respectively
    -0.54
     indicating
    -0.54
     evidently
    -0.54
    POSITIVE LOGITS
     yourself
    0.94
     yourselves
    0.85
     ourselves
    0.85
    fuck
    0.82
     oneself
    0.81
     myself
    0.80
    shit
    0.78
    poke
    0.76
    pipe
    0.74
     smack
    0.69
    Activations Density 0.315%

    No Known Activations