© 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. 22373
    Prev
    Next
    INDEX
    Explanations

    phrases suggesting speculation, opinions, or commentary on various issues

    oai_token-act-pair · gpt-4o-miniTriggered 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
    447
    +0.18
    0.8%
    377
    +0.12
    0.6%
    174
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    157
    +0.18
    0.05
    455
    +0.12
    0.05
    377
    +0.10
    0.05
    Negative Logits
     Shutterstock
    -0.82
     Continuous
    -0.71
    -+
    -0.68
     Zamb
    -0.67
     Delicious
    -0.65
     Saturdays
    -0.65
     Const
    -0.64
     Opening
    -0.64
     Notting
    -0.64
     Deliver
    -0.64
    POSITIVE LOGITS
     speculate
    1.34
     speculated
    1.24
     argue
    1.14
     accuse
    1.10
     criticize
    1.07
     hypothes
    1.04
     theor
    1.00
     wondered
    0.95
     criticized
    0.94
     errone
    0.93
    Activations Density 0.232%

    No Known Activations