© 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. 4197
    Prev
    Next
    INDEX
    Explanations

    words related to behavior and actions, particularly focusing on how individuals conduct themselves in various situations

    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
    +0.13
    0.6%
    288
    +0.11
    0.5%
    665
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.13
    0.03
    70
    +0.11
    0.03
    540
    +0.10
    0.03
    Negative Logits
    andel
    -0.75
    ondo
    -0.67
    corn
    -0.66
    odi
    -0.65
     corpus
    -0.63
     landing
    -0.62
     Destination
    -0.60
     ending
    -0.60
    plet
    -0.60
     Gale
    -0.60
    POSITIVE LOGITS
     differently
    1.17
     similarly
    1.04
     poorly
    0.95
     accordingly
    0.92
     badly
    0.92
     favorably
    0.91
     err
    0.87
    uate
    0.87
     tremend
    0.87
     horribly
    0.86
    Activations Density 0.079%

    No Known Activations