© 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. 4357
    Prev
    Next
    INDEX
    Explanations

    words related to series or episodes in a program or movie franchise

    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
    225
    +0.11
    0.5%
    127
    +0.09
    0.4%
    451
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    651
    +0.11
    0.04
    455
    +0.09
    0.04
    540
    +0.09
    0.04
    Negative Logits
     Places
    -0.64
    verbs
    -0.61
     poles
    -0.59
    robe
    -0.59
     clipboard
    -0.58
    inav
    -0.57
    animate
    -0.57
     limbs
    -0.56
     Horses
    -0.55
    gencies
    -0.53
    POSITIVE LOGITS
     of
    1.17
     thereof
    1.01
    OF
    0.90
     installment
    0.84
     OF
    0.79
     iteration
    0.77
    of
    0.69
     Of
    0.68
    stage
    0.68
     onwards
    0.68
    Activations Density 0.143%

    No Known Activations