© 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. 13808
    Prev
    Next
    INDEX
    Explanations

    phrases related to political conflicts or controversial statements

    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
    571
    +0.15
    0.7%
    516
    +0.15
    0.7%
    98
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    571
    +0.15
    0.04
    98
    +0.15
    0.05
    516
    +0.13
    0.04
    Negative Logits
    ictionary
    -0.68
     Dictionary
    -0.65
     IB
    -0.65
     quotation
    -0.64
     Invention
    -0.62
    BSD
    -0.60
     Funny
    -0.60
     proverb
    -0.59
     Saying
    -0.59
    OSP
    -0.59
    POSITIVE LOGITS
    eded
    0.82
    anium
    0.77
     ever
    0.75
    ammed
    0.72
     someday
    0.70
    rend
    0.69
    utm
    0.68
     dared
    0.67
     succeed
    0.66
     continued
    0.65
    Activations Density 0.152%

    No Known Activations