© 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. 2783
    Prev
    Next
    INDEX
    Explanations

    words related to incidents or events affecting a structure

    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
    251
    +0.12
    0.5%
    616
    +0.11
    0.5%
    85
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    251
    +0.12
    0.02
    12
    +0.11
    0.02
    85
    +0.10
    0.02
    Negative Logits
    buster
    -0.81
    hyde
    -0.73
    journal
    -0.69
     Hunters
    -0.67
     Unch
    -0.65
     Solitaire
    -0.61
    ĸļ
    -0.61
     cleaner
    -0.60
    masters
    -0.60
     Mothers
    -0.60
    POSITIVE LOGITS
    ursions
    1.35
    ursion
    1.32
    idences
    1.31
    isive
    1.30
    ision
    1.29
    itement
    1.25
    ipient
    1.25
    identally
    1.24
    isions
    1.23
    orporated
    1.22
    Activations Density 0.011%

    No Known Activations