© 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 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. 12541
    Prev
    Next
    INDEX
    Explanations

    negative terms associated with politics and conflict

    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
    116
    +0.11
    0.5%
    528
    +0.10
    0.5%
    100
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    116
    +0.11
    0.04
    762
    +0.10
    0.06
    100
    +0.10
    0.04
    Negative Logits
    ispers
    -0.69
    mares
    -0.67
     Administ
    -0.66
    asio
    -0.66
    rises
    -0.64
     Billboard
    -0.62
    edin
    -0.61
    INS
    -0.60
    ecycle
    -0.60
     VIDE
    -0.59
    POSITIVE LOGITS
     unworthy
    1.02
     deserving
    0.91
     undermin
    0.88
     evils
    0.81
     worthy
    0.76
     because
    0.70
     unacceptable
    0.69
     inferior
    0.69
    .[
    0.68
    fascist
    0.68
    Activations Density 0.340%

    No Known Activations