© 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. 5945
    Prev
    Next
    INDEX
    Explanations

    opinions and discussions on various topics

    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
    373
    +0.32
    1.7%
    87
    +0.17
    0.9%
    481
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.32
    -0.03
    8
    +0.17
    0.06
    287
    +0.12
    0.04
    Negative Logits
     Pwr
    -0.76
    elight
    -0.60
    ouses
    -0.56
     Later
    -0.56
    ktop
    -0.56
     thereafter
    -0.55
    sequently
    -0.54
    igree
    -0.53
     afterwards
    -0.53
    eaturing
    -0.53
    POSITIVE LOGITS
     overest
    0.62
     ALWAYS
    0.61
     fundamentally
    0.60
     objectively
    0.60
     millennials
    0.59
     feminism
    0.58
     atheists
    0.58
     fucking
    0.57
     psychologists
    0.56
     libertarians
    0.56
    Activations Density 1.033%

    No Known Activations