© 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. 17234
    Prev
    Next
    INDEX
    Explanations

    strong emotional or descriptive words

    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
    680
    +0.11
    0.5%
    167
    +0.09
    0.4%
    73
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    404
    +0.11
    0.03
    711
    +0.09
    0.02
    288
    +0.09
    0.04
    Negative Logits
    bots
    -0.77
    chairs
    -0.76
    bows
    -0.75
    =~
    -0.74
     Committees
    -0.70
    alysis
    -0.69
     Pets
    -0.68
     shields
    -0.67
    naires
    -0.66
     Contracts
    -0.66
    POSITIVE LOGITS
     piece
    1.08
     thing
    1.02
     item
    0.98
     outing
    0.95
     ounce
    0.94
     element
    0.94
     contender
    0.93
     hurdle
    0.90
     staffer
    0.89
     strand
    0.89
    Activations Density 0.244%

    No Known Activations