© 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. 15728
    Prev
    Next
    INDEX
    Explanations

    information about historical figures aiding military and political efforts

    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
    327
    +0.10
    0.4%
    332
    +0.09
    0.4%
    108
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    327
    +0.10
    0.04
    504
    +0.09
    0.03
    108
    +0.09
    0.03
    Negative Logits
    theless
    -0.75
    course
    -0.69
    meat
    -0.66
    posure
    -0.66
    requisite
    -0.63
    rencies
    -0.62
    Compar
    -0.60
     shenan
    -0.59
    iu
    -0.59
     elig
    -0.57
    POSITIVE LOGITS
     organize
    0.94
     facilitate
    0.93
     organise
    0.93
    fully
    0.91
     usher
    0.87
     oversee
    0.86
    uggle
    0.83
     stabilize
    0.82
     propel
    0.82
     define
    0.82
    Activations Density 0.051%

    No Known Activations