© 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 EvalsExportsinterp-engineNEWAPI 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. 15741
    Prev
    Next
    INDEX
    Explanations

    references to legal and government-related terms

    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
    745
    +0.13
    0.6%
    368
    +0.12
    0.5%
    456
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    718
    +0.13
    0.08
    566
    +0.12
    0.07
    368
    +0.12
    0.06
    Negative Logits
     compat
    -0.73
     predecessor
    -0.64
    izont
    -0.61
     Convers
    -0.60
    rossover
    -0.60
    icipated
    -0.58
     predecessors
    -0.58
     transformations
    -0.58
    fter
    -0.57
     Ascension
    -0.57
    POSITIVE LOGITS
     intervene
    1.08
     swoop
    0.91
     dictate
    0.90
     sanction
    0.88
     frown
    0.87
     relent
    0.86
     incentiv
    0.86
     intervened
    0.85
     kindly
    0.82
     forbid
    0.81
    Activations Density 0.531%

    No Known Activations