© 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 EvalsExportsAPI 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. 11025
    Prev
    Next
    INDEX
    Explanations

    people's names and actions related to activities or events

    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
    480
    +0.51
    2.8%
    447
    +0.18
    1.0%
    266
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.51
    0.03
    266
    +0.18
    0.09
    618
    +0.16
    0.07
    Negative Logits
     deval
    -0.64
     incon
    -0.57
     equival
    -0.55
     meas
    -0.55
     disregard
    -0.55
    Statistics
    -0.55
     deterrent
    -0.55
     multiplied
    -0.54
    fw
    -0.54
     disag
    -0.54
    POSITIVE LOGITS
     guitarist
    0.79
     Photographer
    0.70
     Architects
    0.69
    bass
    0.65
     Motorsport
    0.65
     Cycling
    0.64
     Managing
    0.63
     drummer
    0.63
     Leica
    0.63
     Editor
    0.62
    Activations Density 0.890%

    No Known Activations