© 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. 12643
    Prev
    Next
    INDEX
    Explanations

    descriptions related to a process leading to a crafted or finished result

    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
    459
    +0.10
    0.5%
    174
    +0.09
    0.4%
    71
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    16
    +0.10
    0.02
    122
    +0.09
    0.02
    204
    +0.09
    0.02
    Negative Logits
    hare
    -0.80
    edia
    -0.77
    uli
    -0.75
    uld
    -0.68
    andowski
    -0.65
    orum
    -0.63
    bi
    -0.62
    æ©Ł
    -0.62
    bies
    -0.62
    usions
    -0.61
    POSITIVE LOGITS
     touches
    0.89
     product
    0.78
    strap
    0.74
    ners
    0.69
    work
    0.66
    nings
    0.66
     goods
    0.65
     unfinished
    0.65
     Finish
    0.65
    wall
    0.65
    Activations Density 0.021%

    No Known Activations