© 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. 4724
    Prev
    Next
    INDEX
    Explanations

    dates and numerical figures

    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
    678
    +0.11
    0.5%
    9
    +0.11
    0.5%
    581
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.07
    9
    +0.11
    0.03
    76
    +0.10
    0.04
    Negative Logits
     Lanka
    -0.72
     millenn
    -0.68
     horizont
    -0.62
    Ñı
    -0.58
     è£ıè
    -0.58
    éĹĺ
    -0.57
    Tokens
    -0.57
     Babel
    -0.56
    TextColor
    -0.56
    hap
    -0.55
    POSITIVE LOGITS
     deserves
    0.76
     deems
    0.71
     receives
    0.71
     \'
    0.71
    "]=>
    0.68
    ="#
    0.68
     existed
    0.67
     ensued
    0.65
     opted
    0.63
     resides
    0.63
    Activations Density 0.604%

    No Known Activations