© 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. 17482
    Prev
    Next
    INDEX
    Explanations

    names of individuals

    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
    659
    +0.21
    1.0%
    499
    +0.10
    0.5%
    186
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.21
    0.05
    499
    +0.10
    0.03
    186
    +0.10
    0.03
    Negative Logits
     galleries
    -0.69
     pony
    -0.66
    ãĤ¤ãĥĪ
    -0.66
    holders
    -0.64
     MacArthur
    -0.64
     beware
    -0.62
     captcha
    -0.62
    ciation
    -0.61
     Starfleet
    -0.60
    houses
    -0.60
    POSITIVE LOGITS
    upiter
    1.37
    ournals
    1.29
    ealous
    1.25
    acket
    1.24
    unction
    1.23
    umping
    1.22
    igsaw
    1.22
    ourney
    1.20
    umbo
    1.20
    AMES
    1.19
    Activations Density 0.024%

    No Known Activations