© 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. 4211
    Prev
    Next
    INDEX
    Explanations

    words related to legal terms and proceedings

    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
    102
    +0.13
    0.6%
    278
    +0.10
    0.5%
    627
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.13
    0.10
    55
    +0.10
    0.09
    566
    +0.10
    0.06
    Negative Logits
     strokes
    -0.91
     cubes
    -0.76
    ipeg
    -0.76
     curls
    -0.75
     Stones
    -0.74
     selfies
    -0.74
     highs
    -0.72
     stroke
    -0.70
     Stuff
    -0.70
    phies
    -0.70
    POSITIVE LOGITS
     headquartered
    1.14
     disbanded
    1.08
     complicit
    1.05
     staffed
    1.02
     sponsoring
    1.02
     embroiled
    1.00
     negligent
    0.96
     responsible
    0.94
     cooperating
    0.93
     accountable
    0.92
    Activations Density 0.411%

    No Known Activations