© 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. 14512
    Prev
    Next
    INDEX
    Explanations

    the word "like" followed by a numerical value, often relating to comparisons or similarities

    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
    373
    +0.11
    0.5%
    231
    +0.09
    0.4%
    41
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    572
    +0.11
    0.03
    546
    +0.09
    0.03
    231
    +0.09
    0.03
    Negative Logits
    ourse
    -0.73
    utenberg
    -0.70
    Published
    -0.69
    aptic
    -0.68
    erial
    -0.67
    esson
    -0.67
    endiary
    -0.65
    ulty
    -0.65
    ionics
    -0.64
    hiba
    -0.63
    POSITIVE LOGITS
    lihood
    1.60
     ours
    1.37
    liest
    1.11
     yours
    1.07
    lier
    1.06
     hers
    1.04
     theirs
    1.03
     minded
    0.95
    minded
    0.82
    liness
    0.76
    Activations Density 0.068%

    No Known Activations