© 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. 18867
    Prev
    Next
    INDEX
    Explanations

    information about governmental programs, policies, and community responses to events and issues

    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
    87
    +0.27
    1.4%
    458
    +0.10
    0.5%
    752
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.27
    -0.03
    729
    +0.10
    0.06
    225
    +0.09
    0.07
    Negative Logits
     Shutterstock
    -0.63
    )?
    -0.55
    ());
    -0.53
    ishable
    -0.52
     Himself
    -0.52
    lopp
    -0.50
    uxe
    -0.49
     himself
    -0.49
     lux
    -0.49
     hy
    -0.49
    POSITIVE LOGITS
     regarding
    1.01
     concerning
    0.87
     recommending
    0.86
     approving
    0.82
     authorizing
    0.81
     advising
    0.77
     Regarding
    0.76
     urging
    0.76
     advocating
    0.76
     condemning
    0.75
    Activations Density 1.117%

    No Known Activations