© 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. 1490
    Prev
    Next
    INDEX
    Explanations

    words related to political protests and economic inequality

    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
    87
    +0.30
    1.5%
    756
    +0.12
    0.6%
    268
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.30
    -0.02
    756
    +0.12
    -0.01
    86
    +0.09
    0.05
    Negative Logits
    poke
    -0.73
    Python
    -0.71
    ocular
    -0.69
     nods
    -0.66
     Correct
    -0.64
     Cannes
    -0.63
    iple
    -0.63
    afort
    -0.61
    Marvel
    -0.60
     PowerShell
    -0.60
    POSITIVE LOGITS
     impover
    1.18
     malnutrition
    1.15
     poverty
    1.13
     starvation
    1.13
     worsening
    1.12
     epidemic
    1.12
     worsened
    1.10
     untreated
    1.09
     exacerb
    1.07
     exacerbated
    1.06
    Activations Density 1.163%

    No Known Activations