© 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. 2365
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Acting" followed by a title or position

    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
    339
    +0.13
    0.6%
    373
    +0.11
    0.5%
    685
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    339
    +0.13
    0.01
    111
    +0.11
    0.01
    229
    +0.09
    0.01
    Negative Logits
    xual
    -0.72
    roe
    -0.71
    phia
    -0.66
    joy
    -0.65
    gart
    -0.65
    paralle
    -0.65
    ican
    -0.64
    cutting
    -0.64
    phal
    -0.63
    bringing
    -0.62
    POSITIVE LOGITS
    uary
    0.98
    Director
    0.91
     director
    0.87
     PRESIDENT
    0.84
     Director
    0.84
     prov
    0.84
    uate
    0.80
     administrator
    0.79
     adm
    0.79
     secretary
    0.78
    Activations Density 0.013%

    No Known Activations