© 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. 16482
    Prev
    Next
    INDEX
    Explanations

    phrases related to political and social issues, especially focusing on global or regional events

    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
    468
    +0.13
    0.6%
    450
    +0.12
    0.6%
    364
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    108
    +0.13
    0.04
    468
    +0.12
    0.04
    54
    +0.11
    0.04
    Negative Logits
     Deity
    -0.69
     Donation
    -0.66
     Carib
    -0.62
    acci
    -0.61
    icidal
    -0.60
     sqor
    -0.59
     Byte
    -0.57
     EntityItem
    -0.56
     Cause
    -0.55
    neau
    -0.55
    POSITIVE LOGITS
    adays
    1.83
    here
    1.12
    abouts
    0.75
    IRE
    0.73
     mostly
    0.69
    IPS
    0.65
    heres
    0.64
    mington
    0.63
    nir
    0.63
     retirees
    0.61
    Activations Density 0.100%

    No Known Activations