© 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. 14607
    Prev
    Next
    INDEX
    Explanations

    instances of the word "frog"

    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
    447
    +0.14
    0.7%
    322
    +0.11
    0.5%
    326
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.14
    0.02
    288
    +0.11
    0.03
    500
    +0.11
    0.02
    Negative Logits
    iners
    -0.75
    atars
    -0.73
     concess
    -0.71
    ider
    -0.71
    iam
    -0.71
     Chamberlain
    -0.65
    naire
    -0.64
    isions
    -0.63
    ardi
    -0.63
     Goff
    -0.62
    POSITIVE LOGITS
    lication
    0.80
    metry
    0.79
    mons
    0.71
    bread
    0.66
    Hunt
    0.64
    nil
    0.64
    ERSON
    0.64
     Akin
    0.63
    Prev
    0.63
    Ĥ¬
    0.62
    Activations Density 0.054%

    No Known Activations