© 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. 15825
    Prev
    Next
    INDEX
    Explanations

    instances where authors are cited, specifically looking for the string "et al" and "et al."

    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
    23
    +0.12
    0.5%
    603
    +0.11
    0.5%
    651
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    603
    +0.12
    0.01
    23
    +0.11
    0.01
    651
    +0.11
    0.01
    Negative Logits
    OUNT
    -0.65
    STAR
    -0.64
    canon
    -0.64
    FUL
    -0.61
    IFT
    -0.59
    é¾įå¥ij士
    -0.59
    rontal
    -0.59
     Wage
    -0.59
    theless
    -0.59
    ATTLE
    -0.58
    POSITIVE LOGITS
     seq
    1.28
     al
    1.00
    rics
    0.96
    ches
    0.90
     Associates
    0.89
    rix
    0.86
    iam
    0.86
    ree
    0.86
    jee
    0.84
    sis
    0.83
    Activations Density 0.007%

    No Known Activations