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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 18868
    Prev
    Next
    INDEX
    Explanations

    physical combat and martial arts-related terms

    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
    314
    +0.16
    0.7%
    481
    +0.13
    0.6%
    607
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.16
    0.04
    288
    +0.13
    0.05
    500
    +0.10
    0.04
    Negative Logits
    vae
    -0.79
     Monstrous
    -0.70
    aples
    -0.69
     reconc
    -0.68
     KDE
    -0.68
     Shiite
    -0.67
     Fiscal
    -0.65
     Pione
    -0.65
     Provincial
    -0.65
     Jub
    -0.65
    POSITIVE LOGITS
    able
    1.26
    ings
    1.17
    ability
    1.08
    ers
    1.02
    pipe
    1.01
    tor
    0.99
    masters
    0.98
    cart
    0.98
    athon
    0.98
    stick
    0.98
    Activations Density 0.186%

    No Known Activations