© 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. 10154
    Prev
    Next
    INDEX
    Explanations

    words related to high-profile political figures and media coverage

    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
    675
    +0.10
    0.5%
    314
    +0.10
    0.5%
    23
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    675
    +0.10
    0.10
    759
    +0.10
    0.09
    514
    +0.10
    0.08
    Negative Logits
     guiActiveUnfocused
    -0.78
     democracies
    -0.67
     MAP
    -0.61
     battleground
    -0.60
     disadvant
    -0.60
     democracy
    -0.59
     discrep
    -0.59
     Democracy
    -0.58
     Palestin
    -0.58
     Mush
    -0.58
    POSITIVE LOGITS
    own
    0.84
    felt
    0.83
    tarian
    0.82
    ¯
    0.76
    ï¸ı
    0.75
    iated
    0.74
    ves
    0.74
    ufact
    0.74
     knows
    0.73
    bart
    0.73
    Activations Density 0.252%

    No Known Activations