© 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. 12798
    Prev
    Next
    INDEX
    Explanations

    phrases related to taking action or making a statement

    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
    324
    +0.10
    0.5%
    608
    +0.10
    0.5%
    665
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.10
    0.05
    608
    +0.10
    0.05
    718
    +0.09
    0.04
    Negative Logits
    opter
    -0.72
    Az
    -0.68
    Ox
    -0.66
    ophy
    -0.66
    yd
    -0.65
     existed
    -0.65
    ounter
    -0.62
    omes
    -0.62
     Knot
    -0.61
    Stars
    -0.60
    POSITIVE LOGITS
     blackmail
    1.02
     leverage
    1.00
     pretext
    0.94
     motivation
    0.87
     justify
    0.87
     excuse
    0.87
     shove
    0.86
     rallying
    0.82
     motivate
    0.82
     advantage
    0.82
    Activations Density 0.223%

    No Known Activations