© 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. 763
    Prev
    Next
    INDEX
    Explanations

    references to specific individuals, locations, and events in news articles

    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
    430
    +0.14
    0.6%
    481
    +0.13
    0.6%
    470
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    765
    +0.14
    0.10
    430
    +0.13
    0.14
    452
    +0.11
    0.08
    Negative Logits
    Conclusion
    -0.81
     etc
    -0.73
    usercontent
    -0.71
    itized
    -0.71
    })
    -0.70
    ĪĴ
    -0.70
     blah
    -0.68
     forg
    -0.67
    CONCLUS
    -0.65
     âĨij
    -0.63
    POSITIVE LOGITS
     reportedly
    0.92
     flanked
    0.89
     unveiled
    0.84
     hailed
    0.81
     dubbed
    0.80
     originally
    0.79
     posted
    0.76
     debuted
    0.75
     billed
    0.74
     penned
    0.73
    Activations Density 3.080%

    No Known Activations