© 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. 24370
    Prev
    Next
    INDEX
    Explanations

    references to art or artistic expression

    oai_token-act-pair · gpt-4o-miniTriggered 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
    314
    +0.10
    0.5%
    14
    +0.10
    0.5%
    167
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    523
    +0.10
    0.01
    334
    +0.10
    0.01
    670
    +0.10
    0.01
    Negative Logits
     htt
    -0.78
     sclerosis
    -0.66
     Telecommunications
    -0.61
     Contra
    -0.61
     schedules
    -0.60
     carcin
    -0.60
     iod
    -0.59
    kHz
    -0.59
     schedule
    -0.59
    KI
    -0.58
    POSITIVE LOGITS
    illery
    1.51
    ifacts
    1.49
    isan
    1.47
    ificial
    1.39
    isans
    1.33
    icles
    1.27
    emis
    1.26
    ifact
    1.22
    esian
    1.21
    ific
    1.15
    Activations Density 0.012%

    No Known Activations