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

    Neuronpedia

    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 3072
    9. 8-RES_FS3072-JB
    11. 105
    Prev
    Next
    INDEX
    Explanations

    actions and events related to law enforcement, legal terms, and police activities

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    168
    +0.12
    0.6%
    104
    +0.10
    0.5%
    217
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    469
    +0.12
    0.09
    168
    +0.10
    0.09
    665
    +0.09
    0.10
    Negative Logits
    ollah
    -0.82
    RH
    -0.74
    margin
    -0.66
    dp
    -0.64
    ahime
    -0.61
     conglomer
    -0.61
     derivatives
    -0.60
     subp
    -0.59
     reliance
    -0.59
     ramifications
    -0.59
    POSITIVE LOGITS
     voluntarily
    0.89
     hither
    0.88
     aboard
    0.80
     town
    0.77
     alive
    0.76
     downstairs
    0.74
     doorstep
    0.74
     unnoticed
    0.73
     porch
    0.73
     anytime
    0.69
    Activations Density 1.353%

    No Known Activations