© 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. 18793
    Prev
    Next
    INDEX
    Explanations

    words related to legal issues, law enforcement, and government announcements

    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
    373
    +0.10
    0.5%
    756
    +0.10
    0.4%
    364
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.10
    0.04
    659
    +0.10
    0.03
    284
    +0.10
    0.03
    Negative Logits
    ktop
    -0.77
     msec
    -0.65
     agre
    -0.64
     SPD
    -0.64
     Revelations
    -0.64
     sack
    -0.62
     Redux
    -0.61
     mileage
    -0.60
     converge
    -0.59
     conclud
    -0.58
    POSITIVE LOGITS
    hao
    1.16
    omi
    1.15
    ept
    1.06
    hua
    1.05
    wei
    1.05
     Wei
    1.00
     Lei
    1.00
    jiang
    1.00
    Yu
    0.99
    xi
    0.99
    Activations Density 0.072%

    No Known Activations