© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 2011
    Prev
    Next
    INDEX
    Explanations

    statements related to social justice

    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
    86
    +0.15
    0.7%
    759
    +0.13
    0.6%
    241
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    86
    +0.15
    0.06
    762
    +0.13
    0.06
    431
    +0.10
    0.04
    Negative Logits
     disclaimer
    -0.81
    Quant
    -0.79
     Courier
    -0.73
     Inventory
    -0.73
     Canaver
    -0.71
     Technique
    -0.71
     Heist
    -0.71
    Aim
    -0.69
     Pair
    -0.68
     Correct
    -0.68
    POSITIVE LOGITS
     oppressed
    1.28
     victimized
    1.20
     suffering
    1.13
     tragically
    1.10
     sacrificed
    1.10
     perished
    1.10
     marginalized
    1.09
     suffer
    1.08
     harmed
    1.08
     impacted
    1.04
    Activations Density 0.428%

    No Known Activations