© 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 49152
    9. 8-RES_FS49152-JB
    11. 41652
    Prev
    Next
    INDEX
    Explanations

    words related to the concept of being original or initial

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    +0.10
    0.5%
    99
    +0.09
    0.4%
    738
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    532
    +0.10
    0.02
    738
    +0.09
    0.02
    99
    +0.09
    0.02
    Negative Logits
    rom
    -0.87
    =-=-=-=-=-=-=-=-
    -0.77
    cig
    -0.69
    ritch
    -0.66
    rament
    -0.64
    eer
    -0.63
    holes
    -0.62
    adel
    -0.62
    rw
    -0.62
    lov
    -0.61
    POSITIVE LOGITS
     intended
    1.14
     planned
    1.05
     conceived
    1.02
     slated
    0.97
     envisioned
    0.95
     scheduled
    0.88
     designed
    0.83
     assigned
    0.82
     hoped
    0.82
     belonged
    0.79
    Activations Density 0.017%

    No Known Activations