© 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. 7478
    Prev
    Next
    INDEX
    Explanations

    terms related to trademark law and copyright protection

    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
    589
    +0.13
    0.6%
    420
    +0.12
    0.5%
    390
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    555
    +0.13
    0.02
    588
    +0.12
    0.02
    288
    +0.12
    0.02
    Negative Logits
     assassination
    -0.72
     straw
    -0.66
    pine
    -0.64
    gur
    -0.63
     interrogation
    -0.62
     slaughter
    -0.62
     combustion
    -0.59
     delivery
    -0.58
     erection
    -0.58
     ruining
    -0.58
    POSITIVE LOGITS
    itionally
    1.61
    emark
    1.27
    ename
    1.25
    eless
    1.22
    elines
    1.15
    ework
    1.14
    etooth
    1.13
    eways
    1.13
    itional
    1.11
    emen
    1.08
    Activations Density 0.040%

    No Known Activations