© 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. 15867
    Prev
    Next
    INDEX
    Explanations

    words related to historical events or stories

    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
    224
    +0.12
    0.5%
    393
    +0.11
    0.5%
    547
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    224
    +0.12
    0.04
    149
    +0.11
    0.04
    430
    +0.09
    0.04
    Negative Logits
     misogyn
    -0.68
     economical
    -0.65
    ³³³³
    -0.65
    Georg
    -0.64
    destruct
    -0.63
    urg
    -0.63
     arbitrary
    -0.63
    interstitial
    -0.63
     lethal
    -0.60
     unacceptable
    -0.60
    POSITIVE LOGITS
     oversaw
    1.27
     oversees
    1.24
     owns
    1.12
     specializes
    1.11
     participated
    1.10
     attended
    1.10
     attends
    1.10
     specialize
    1.06
     chaired
    1.03
     authored
    1.02
    Activations Density 0.095%

    No Known Activations