© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 2950
    Prev
    Next
    INDEX
    Explanations

    phrases related to debates and political matters

    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
    314
    +0.12
    0.6%
    403
    +0.10
    0.5%
    371
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    371
    +0.12
    0.07
    467
    +0.10
    0.06
    382
    +0.10
    0.05
    Negative Logits
    Russ
    -0.61
    fill
    -0.61
    FTWARE
    -0.60
    HTTP
    -0.60
    Pac
    -0.58
    available
    -0.57
    cro
    -0.56
    ships
    -0.56
    bage
    -0.56
    interrupted
    -0.55
    POSITIVE LOGITS
     least
    1.14
    abase
    0.99
    yp
    0.93
    onement
    0.91
    roph
    0.91
     home
    0.87
     halftime
    0.83
     times
    0.82
     stake
    0.81
    mosp
    0.81
    Activations Density 0.186%

    No Known Activations