© 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. 5064
    Prev
    Next
    INDEX
    Explanations

    sentences or phrases that end with exclamations or statements emphatically

    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
    447
    +0.30
    1.5%
    640
    +0.23
    1.1%
    480
    +0.22
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.30
    0.09
    640
    +0.23
    0.08
    679
    +0.22
    0.07
    Negative Logits
     withd
    -0.77
     encount
    -0.74
     advoc
    -0.73
     surpr
    -0.71
    onga
    -0.70
     stip
    -0.69
     redes
    -0.68
     glim
    -0.68
     challeng
    -0.64
     frontline
    -0.63
    POSITIVE LOGITS
    txt
    0.97
    jpg
    0.97
     ;)
    0.92
     :)
    0.92
     Please
    0.91
     Literally
    0.90
    tumblr
    0.88
     You
    0.85
    gif
    0.85
     Especially
    0.84
    Activations Density 0.247%

    No Known Activations