© 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. 12934
    Prev
    Next
    INDEX
    Explanations

    proper names

    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
    525
    +0.14
    0.7%
    751
    +0.13
    0.6%
    377
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.14
    0.04
    659
    +0.13
    0.03
    701
    +0.11
    0.03
    Negative Logits
    ngth
    -0.67
     reven
    -0.61
     spoilers
    -0.59
     Uriel
    -0.58
     tremend
    -0.56
    igible
    -0.54
    ennes
    -0.54
    ertation
    -0.53
    endo
    -0.51
    marked
    -0.51
    POSITIVE LOGITS
    ĵĺ
    0.76
    reau
    0.75
    hemy
    0.73
    anchester
    0.73
    deals
    0.71
    angles
    0.68
     Wales
    0.67
    amia
    0.67
    swick
    0.63
     Garland
    0.63
    Activations Density 0.126%

    No Known Activations