© 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. 5054
    Prev
    Next
    INDEX
    Explanations

    strings of words that are repeated multiple times with a high level of similarity

    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
    167
    +0.12
    0.6%
    587
    +0.12
    0.6%
    678
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    587
    +0.12
    0.01
    129
    +0.12
    0.01
    180
    +0.10
    0.01
    Negative Logits
    tical
    -0.80
    undai
    -0.77
    icago
    -0.73
    hammad
    -0.71
    espie
    -0.67
     Flavoring
    -0.66
    ternity
    -0.64
    hemat
    -0.64
    yrim
    -0.63
    vantage
    -0.63
    POSITIVE LOGITS
     bikini
    0.93
    ency
    0.91
    ently
    0.84
    encies
    0.83
    inous
    0.78
    entially
    0.77
    angled
    0.76
    angle
    0.75
    asso
    0.73
     Cheese
    0.72
    Activations Density 0.017%

    No Known Activations