© 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. 16363
    Prev
    Next
    INDEX
    Explanations

    action verbs related to the evaluation or analysis of events or people

    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
    551
    +0.11
    0.5%
    440
    +0.10
    0.5%
    541
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.11
    0.04
    540
    +0.10
    0.04
    217
    +0.09
    0.03
    Negative Logits
    letters
    -0.64
    adra
    -0.63
    ede
    -0.62
    Layout
    -0.61
    edes
    -0.61
    Torrent
    -0.60
    hold
    -0.60
     Tale
    -0.59
     Clicker
    -0.56
    Found
    -0.56
    POSITIVE LOGITS
     separately
    1.14
     individually
    1.05
    ĸļ
    1.00
     extensively
    0.97
     thoroughly
    0.94
     anew
    0.92
     favorably
    0.91
     psychiat
    0.87
     by
    0.86
    igated
    0.84
    Activations Density 0.129%

    No Known Activations