© 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. 2432
    Prev
    Next
    INDEX
    Explanations

    mentions of societal and political issues

    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
    481
    +0.10
    0.5%
    187
    +0.10
    0.5%
    737
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    86
    +0.10
    0.06
    737
    +0.10
    0.05
    187
    +0.10
    0.05
    Negative Logits
    Operation
    -0.84
    catentry
    -0.77
    Accessory
    -0.75
    tnc
    -0.73
    \/\/
    -0.72
     srfAttach
    -0.72
    Solution
    -0.71
    Va
    -0.70
    Emer
    -0.69
    Sensor
    -0.68
    POSITIVE LOGITS
     smugglers
    1.22
    hood
    1.07
     who
    1.04
    folk
    0.99
     traffickers
    0.93
    opausal
    0.89
     whom
    0.86
     smuggling
    0.85
     born
    0.83
    liness
    0.81
    Activations Density 0.131%

    No Known Activations