© 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. 22917
    Prev
    Next
    INDEX
    Explanations

    the occurrence of the word "next" in phrases related to upcoming events or predictions

    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
    223
    +0.10
    0.5%
    373
    +0.10
    0.4%
    530
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    223
    +0.10
    0.04
    530
    +0.10
    0.04
    458
    +0.09
    0.04
    Negative Logits
    lees
    -0.89
    xual
    -0.79
    theless
    -0.77
    lee
    -0.73
     Franch
    -0.67
    tics
    -0.66
    olor
    -0.66
    then
    -0.66
     Lauder
    -0.66
    had
    -0.64
    POSITIVE LOGITS
     generation
    1.12
     installment
    1.08
     iteration
    1.02
     decade
    0.94
     generations
    0.93
     incarnation
    0.93
     phase
    0.91
     step
    0.91
     occupant
    0.90
     batch
    0.89
    Activations Density 0.036%

    No Known Activations