© 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. 24502
    Prev
    Next
    INDEX
    Explanations

    the word "where" indicating locations or contexts

    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
    423
    +0.09
    0.4%
    203
    +0.09
    0.4%
    177
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.09
    0.03
    300
    +0.09
    0.03
    661
    +0.08
    0.02
    Negative Logits
    apult
    -0.86
    TAG
    -0.83
    wik
    -0.66
    mask
    -0.65
    Und
    -0.63
    TED
    -0.63
    format
    -0.62
    yp
    -0.62
    ]);
    -0.61
    Sec
    -0.61
    POSITIVE LOGITS
    upon
    1.22
    abouts
    0.99
     temperatures
    0.85
     winters
    0.83
    soever
    0.82
     homosexuality
    0.78
     mobs
    0.77
     residents
    0.75
     unemployment
    0.75
    fore
    0.74
    Activations Density 0.049%

    No Known Activations